A/A - Active/Active; sometimes called true Active/Active; a load balancing method used to distribute a load among two or more devices. In active/active systems, all paths are active at all times, unless a path fails. Active/active is commonly used for load balancing. Contrast Active/Active (A/A) with Active/Passive (A/P) and Asymmetrical (ALUA).
Active/Active configurations use several different methods: RR, HBD, or SBD.
AA/A - Asymmetric Active/Active Also see VMware.
AAL5/A - ATMA> Adaptation Layer for computer data
A/P - Active/Passive; a method for addressing a disk array whereby each LUN is associated with a preferred controller and all I/O for that LUN is transmitted and received via that controler. An alternative (standby) controller can await instructions to take ownership of the LUN in the eventuality that path failover is invoked. Once failover is invoked, the alternate controll assumes control of the LUN and facilitates reads and writes until control is passed back to the original controller. Active/Passive arrays do not allow true load balancing, but are usually configured to ensure that there is an even spread of activity. In active/passive systems, backup paths are not active until a path fails. Active/passive is commonly used for redundancy. Contrast Active/Passive (A/P) with Active/Active (A/A) and Asymmetrical (ALUA).
ABOD - Storage Application Support JBOD. Contrast with EBOD, JBOD, RBOD, SBOD and JBOK.
ACL - Access Control Lists
ADIC - Advanced Digital Information Corporation; Acquired by Quantum in August 2006.
access fairness - A process by which contending nodes are guaranteed access to an FC-AL.
access method - The means used to access a physical medium in order to transmit data.
ACL - Access Control List: On Windows file systems, an ACL is a list of permissions attached to an object (e.g. file, directory). The ACL specifies which users or system processes are granted access to objects, as well as what operations are allowed on given objects.
active copper - kA fibre channel connection that allows up to 30 meter copper cabling between devices.
Adaptec - Adaptec See http://www.adaptec.com.
address identifier - A 24-bit value used to indicate the link-level address of communicating devices. In a frame header, the address identifier indicates the source identifier (S_ID) and destination identifier (D_ID) of the frame.
AHCI - Advanced Host Controller Interface - Intel technical standard for Serial ATA (SATA) host bus adapters.
AFR - Annualized Failure Rate: A method of measuring failure rates or trends for a group of units at a site. The rates are based on the monthly total number of returned field failure units divided by the total cumulative installed base and multiplied by 12 (to annualize the failure rate).
AL_PA - Arbitrated Loop Physical Address; an 8-bit value used to identify a participating device in an FC-AL.
Alanco Technologies Inc. - http://www.alanco.com/ -
ALUA - Asymmetric Logical Unit Access; When multipath is used in storage (e.g. multiple controllers), the ALUA storage model is often used to ensure that the most efficient path is taken between host and the storage. ALUA is defined in the SPC-3 standard. Contrast Asymmetrical (ALUA) with Active/Active (A/A) and Active/Passive (A/P). Also see VMware.
Amazon EC2 - Amazon Elastic Cloud Compute
Amazon S3 - Amazon Simple Storage Service; See S3.
AMI - American Megatrends; Hardware and software company particularly known for BIOS firmware. See http://www.ami.com.
Amplidata - Large data and cloud storage provider. Their main product is Himalaya and the underlying program is called Distributed Storage System (DSS). In 2015 they were acquired by HGST. See http://www.amplidata.com
Apache - Apache Software Foundation is an American non-profit corporation to support Apache software projects.
AOP - Annual Operating Plan
API - Application Program Interface; Layer of documented services and protocols that separate a resource from the access of that resource. APIs are used to keep an interface distinct from any implementation issues and to localize modifications (i.e. avoid the need to propogate changes throughout a system).
apt-get - Advanced Packaging Tool is the Debian software package manager. See yum and subscription-manager.
Arakoon - Lattus underlying technology for distributed key-value store. One master at a time with a small number of slaves. Almost all client interactions are made with the master. Slaves can be directly contacted in order to identify the master. A new master is elected upon failure. Slaves keep copies of the key-value store and repeated match logs with the master for updates. See http://www.arakoon.org
arbitration - A means for gaining orderly access to a shared-loop topology.
areal density - Magnetic disk recording surface capability usually expressed as terabits per square inch (Tb/in2). Approximate densities predicted are 2015: 1 Tb/in2, 2019: 5 Tb/in2, 2035: 10 Tb/in2.
ARP - Address Resolution Protocol; an IP function that correlates an IP network address to a link-level MAC address.
ARR - Annual Replacement Rate
ASN.1 - Abstract Syntax Notation One: Data definition language which is an international standard for efficient and portable information exchange without regard to how it is represented while in transit.
ASIC - Application-specific integrated circuit.
AT - Advanced Technology (AT); reference to the IBM AT personal computer of 1984.
ATA - AT Attachment (ATA); also called Parallel ATA (PATA); a computer bus interface that connects host bus adapters to mass storage devices such as HDD and optical drives.
ATM - Asynchronous transfer mode; a high-speed cel-switching transport used primarily in wide are networks.
ATTO - ATTO Technology, Inc. with home offices in Amherst, New York. See http://attotech.com/
AWS - Amazon Web Services; The Cloud services offered by the Amazon company.
AXR - Amplidata eXtreme REST - HGST (formerly Amplidata) proprietary storage protocol used by Lattus.
AXR - Nexsan Assureon AXR Archival Storage Appliance. Assureon archive storage systems are very popular in the financial and healthcare industries, federal, state and local government organizations, as well as call centers, video surveillance and organizations with multiple remote offices. Lattus offers an AXR interface API.
Azure - This is Microsoft's cloud service.
bandwidth - Transmission capacity of a link or a system.
BB_credit - Buffer-to-buffer credit; used to determine how many frames can be sent to a recipient. Contrast with EE_Credit.
BGPv4 - Border Gateway Protocol; Used on managed switches like Dell's Force10 S55.
BIOS - Basic Input Output System; Small flash EEPROM memory module on the motherboard of computers that provides the initial settings and basic I/O primitives for boot and other very initial and low level operations.
BitLocker - A utility in Windows, starting with Vista, that encrypts the contents of the hard disk. At startup, the drive must be unlocked by typing in a password or inserting a USB flash drive that contains the password. If a TPM chip is on the motherboard, Bitlocker uses it to generate keys based on system files. If any of these files have changed, BitLocker keeps the data encrypted.
Bit Spread - The policy used to distribute the bits of an entity among disks in an object store.
BitSpread Client Daemon - Lattus service/process/daemon which runs on each controller node.
Blockstore - Lattus logical representation of the disk mount points (ext4 partitions) on which the encoded data is stored.
Blockstore - Baseboard Management Controller; Microcontroller embedded on the motherboard of a computer, generally a server, that manages the interface between system management software and platform hardware. The BMC provides the IPMI architecture. See IPMI.
BPMR - Bit Patterned Magnetic Recording with SMR and TDMR.
Brocade Communications Systems Inc. See http://www.brocade.com/
Bug Scrub - Meeting to discuss the status of reported problems (i.e. bugs).
bypass circuitry - Circuits that automatically remove a device from the data path when valid signaling is lost.
C10 - Lattus Controller Node (Dell R620)
C5 - Lattus Controller Node (Dell R320)
Cache Daemon - Lattus daemon that caches superblocks (data) for frequently read objects.
CAM - Content Addressable Memory
Caml - Categorical Abstract Machine Language. See INRIA
CAGR - Cumulative Annual Growth Rate; This is often applied to the increase in performance or density of disks.
CAPTCHA - Completely Automated Public Turing test to tell Computers and Humans Apart - A test used to make sure that a human is using a system, not a computer program. The test typically involves reading distorted text.
carve the LUN - Partition an area in order to put metadata and data onto the same logical unit.
cascade - Connecting two or more fibre channel hubs or fabric switches to increase the number of ports or to extend distances.
ccNUMA - Cache Coherent Non-Uniform Memory Access; multi-processor computer memory design where some CPUs have a more direct route to memory than others. The result of this is that memory access time depends on the memory location relative to the processor.
CD - Compact Disk
CDN - Content Delivery Network, helps speed up a web site
CDR - clock and data recovery circuitry.
CentOS - Community Enterprise Operating System; Red Hat's open enterprise operating system
CFI - Customer Factory Integration - Dell's program for providing computers built to order and delivered according to customer specifications directly from the factory.
CFM - Connectivity Fault Management; This is defined by IEEE 802.1ag. Used on managed switches like Dell's Force10 S55.
Ethernet CFM is an end-to-end per-service-instance (per VLAN) Ethernet layer OAM protocol. It includes proactive connectivity monitoring, fault verification, and fault isolation. End-to-end can be provider-edge-to-provider-edge (PE-to-PE) device or customer-edge-to-customer-edge (CE-to-CE) device. Ethernet CFM, as specified by IEEE 802.1ag, is the standard for Layer 2 ping, Layer 2 traceroute, and end-to-end connectivity verification of the Ethernet network.
CIDR - Classless Inter-Domain Routing is a method for assigning IP addresses without using the standard IP address classes like Class A, Class B or Class C. In CIDR notation, an IP address is represented as A.B.C.D /n, where "/n" is called the IP prefix or network prefix. The IP prefix identifies the number of significant bits used to identify a network. For example, 192.9.205.22 /18 means, the first 18 bits are used to represent the network and the remaining 14 bits are used to identify hosts. Common prefixes are 8, 16, 24, and 32.
CIFS - Common Internet File System; Microsoft Windows Server MEssage Block (SMB) application-layer netowrk protocol mainly used for providing shared access to files, printers, serial ports, and miscellaneous communications between nodes on a network. It also provides an authenticated inter-process communication mechanism. See SMB and SAMBA.
CIM - Common Information Model; a management structure enablink disparate resources to be managed by a common application.
Class 1 - For fibre channel, a connection-oriented class of service that requires acknowledgment of frame delivery. Contrast with Class 2, Class 3, Class 4, Class 6.
Class 2 - For fibre channel, a connectionless class of service that requires acknowledgment of frame delivery. Contrast with Class 1, Class 3, Class 4, Class 6.
Class 3 - For fibre channel, a connectionless class of service that requires no notification of frame delivery . Contrast with Class 1, Class 2, Class 4, Class 6.
Class 4 - For fibre channel, a class of service that defines virtual circuits via fractional bandwidth and quality-of-service parameters. Contrast with Class 1, Class 2, Class 3, Class 6.
Class 6 - For fibre channel, a class of service that provides multicast frame delivery with acknowledgment. Contrast with Class 1, Class 3, Class 4, Class 6.
CLI - Command Line Interface; A user interface that accepts and acts upon commands typed into a standard REPL.
Cloud Operating System - The operating system that runs the cloud provider's center for data.
Cloud Storage - Virtual storage where users can save their data on a server.
Cloudware - A type of application software that is responsible for running and managing systems
CLP - Command Line Protocol; See SMASH-CLP
CMR - Conventional Magnetic Recording hard disk drive
Compare with SMR
Note: Acronym also used for Colossal Magnetoresistive Heads
COSBench - Cloud Object Storage Benchmark Service - Open sourced benchmark tool for cloud object storage services.
CPLD - Complex Programmable Logic Device
CRC - Cyclic redundancy check; an error-detection method.
CRM - Customer Relationship Management, a widely implemented model for managing a company's interactions with customers, clients, and sales prospects.
Crossroads Systems Inc. - http://www.crossroads.com/
CRU - Customer Replaceable Unit; A system component expected to be replaced by the customer without service training. See FRU
CSB - Cloud Service Brokerage
CSP - Cloud Service Provider
C-state - Dell power mode to conserve energy. Lower-power modes cut the dock signal and power from idle units inside the CPU. Units are stopped (by cutting the clock) or voltage is reduced or completely shut off. Time is required for the CPU to "wake up" and be operational again. C-states are numbered starting at C0 with higher numbers representing decreasing power consumption. C0 (normal operating state), C1 (halt), C1E (enhanced halt), C2 (stop grant/stop clock), C2E (extended stop grant), C3 (sleep/deep sleep/altVID), C4 (depper sleep), C4E/C5 (enhanced deeper sleep), C6 (deep power down).
CTU - Customer Testable Unit: Manufacturing production maturity such that customers can actually test units.
CVFS - CentraVision File System; Original name of StorNext. Originally created to provide fast data transfer between Windows and SGI's IRIX computers. See MountainGate, ADIC, and Quantum
CYA - Cover Your Ass; Crass description of a statement or act designed to equivocate in order to protect from possible criticism or legal penalties.
cut-through - A switching technique that allows a routing decision to be made as soon as the destination address of a frame is received.
CVFS - The original name of StorNext was CentraVision File System (CVFS) created by MountainGate Imaging systems Corporation to provide fast data fransfer between Windows and SGI's IRIX computers.
data storage devices - Companies in the data storage devices sector include: Alanco Technologies Inc., Brocade Communications Systems Inc., Crossroads Systems Inc., DataDirect Networkds, Datalink Corporation, Dataram Corporation, Emulex Corporation, EMC Corporation Dot Hill systmes Corporation, Imation Corp., Nimble Storage Inc., NetApp Inc., Overland Storage Inc., SanDisk Corp., Seagate Technology Public Limited Company, Teradata Corporation, Violin Memory Inc., and Western Digital Corporation.
DataDirect Networks - DDN - http://www.datadirect.com/
See WOS.
Datalink Corporation - http://www.datalink.com/
DKMS - Dynamic Kernel Module Support; See http://www.dell.com/dkms/dkms-ols2004.pdf
DTR - Data Transfer Rate: The rate that digital data is transferred from one point to another, expressed in bits per second or bytes per second. Data Transfer Rate to Disk: The internal disk transfer rate in Mbits per second. Data Transfer Rate from the Buffer to the Host: Based on the transfer of buffered data in MB per second.
Dataram Corporation - http://www.dataram.com/
Dell - Computer and computer-related product manufacturer and reseller. http://www.dell.com/
DIMM - Dual Inline Memory Module; A DIMM is a small circuit board that holds memory chips. It uses a 64-bit bus to the memory, whereas a single in-line memory module ( disk array - A collection of HDD devices. The HDDs are often arranged in disk drive trays with an associated controller.
Disk Safety - Lattus number of Blockstores from a specific spread that can be lost while still being able to recover the original data. If more Blockstores are lost than the Disk Safety allows, the data will no longer be recoverable.
disparity - The relationship of 1s and 0s in an encoded character: positive dispartiy contains more 1s; negative disparity contains more 0s; neutral disparity contains an even number of 1s and 0s.
DLC - Distributed LAN Client proprietary protocol from Quantum.
DMI - Desktop Management Interface table. This table contains a description of the system's hardware components, as well as other useful pieces of information such as serial numbers and BIOS revision. Use the DMIDECODE program to dump. See SMBIOS and DMTP.
dmidecode - Desktop Management Interface table dump program. This is a tool to dump a computer's hardware components, serial numbers and BIOS settings. See DMI.
DMP - Dynamic Multipathing; a method of providing two or more hardware paths to a single drive for high availability and improved performance. See Veritas.
DMRAID - Linux software RAID manipulation command (Try man pages). This is useful to recognize and use vendor RAID solutions.
DMTP - Desktop Management Task Force. Developed DMI and SMBIOS tables.
dnf - yum replacement. Officially stands for nothing. See subscription-manager and apt-get.
Dot Hill Systems Corporation - http://www.dothill.com/
DRBD - Distributed Replicated Block Device
DSA - Dual Stage Actuator: Disk drive improvement to the overall capability of the Servo system and provides a mechanical benefit to improve the response time (higher bandwidth capability) of moving and maintaining the head position over the data tracks.
DSET - Dell Server E-Support Tool: Provides the ability to collect hardware, storage and operating system information from a Dell PowerEdge server.
DSM - Device Specific Module (DSM); a component in MS Windows Server 2008 designed to work with storage arrays that support ALUA controller model as well as storage arrays that follow the AA/A controller model. Microsoft DSM is characterized by the following features:
DSS - Data Security Standard. The acronym is also used for Display Subsystem and Amplidata's Distributed Storage System.
EBOD - Enterprise-class JBOD. Contrast with ABOD, JBOD, RBOD, SBOD and JBOK.
ECC - Error Correction Code: A mathematical algorithm utilizing LDPC technology that can detect and correct errors in a data field by adding check bits to the original data.
ECDSA - Elliptic Curve Digital Signature Algorithm: Digital Signature Algorithm used in security applications.
ECO - Engineering Change Order; Formal change request to be tracked as part of a product development cycle.
EDID - Extended Display Identification Data; This is the information thtat the monitor sends to a PC about itself. Specifically this includes a list of acceptable resolutions.
EE_Credit - End-to-end credit; used to manage the exchange of frames by two communicating devices. Contrast with BB_Credit.
EEE - Energy-Efficient Ethernet; Ethernet device standard that allows for less power consumption during periods of low data activity. This is sometimes called Green Ethernet.
EPROM - Erasable Programmable Read Only Memory; memory that uses floating-gate transistors and can be erased by strong ultraviolet light. Without the erasing light the memory stores data without electrical power. EPROM does not need a battery or other power source to retain data.
EEPROM - Electrically Erasable Programmable Read Only Memory; memory that stores data using electrical charges that maintain their state without electrical power. EEPROM does not need a battery or other power source to retain data. The most common type of EEPROM is flash memory such as used in USB keychain drives.
E_Port - Expansion Port; An expansion port connection two fabric switches. Contrast with F_Port, FL_Port, G_Port, N_Port, NL_Port.
8b/10b encoding - An encoding schemem that converts an 8-bit byte into two possible 10-bit characters; used for balancing 1s and 0s in high-speed transports.
Elastic Cloud Computing - The system automatically distributes the resources to meet the demand.
EMACS - Highly ported and eminently extensible editor and related utilities. EMACS is written is ELISP with C routines for basic support.
EMC - http://emc.com/ - EMC Corporation; Specializes in products for medium to large businesses and is a large provider of data storage systems. Major competitors in data storage systems are HDS. HP, IBM, and NetApp.
EMEA - Europe Middle East and Africa: Albania, Algeria, Andorra, Angola, Austria, Bahrain, Belarus, Belgium, Benin, Bosnia and Herzegovina, Botswana, Bulgaria, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Comoros, Croatia, Cyprus, Czech Republic, Democratic Republic of the Congo, Denmark, Djibouti, Egypt, Equatorial Guinea, Eritrea, Estonia, Ethiopia, Faroe Islands, Finland, France, Gabon, Gambia, Georgia, Germany, Ghana, Gibraltar, Greece, Guernsey, Guinea, Guinea-Bissau, Hungary, Iceland, Iran, Iraq, Ireland, Isle Of Man, Israel, Italy, Ivory Coast, Jersey, Jordan, Kenya, Kuwait, Latvia, Lebanon, Lesotho, Liberia, Libya, Liechtenstein, Lithuania, Luxembourg, Macedonia, Madagascar, Malawi, Mali, Malta, Mauritania, Mauritius, Moldova, Monaco, Montenegro, Morocco, Mozambique, Namibia, Netherlands, Niger, Nigeria, Norway, Oman, Palestine, Poland, Portugal, Qatar, Romania, Rwanda, San Marino, Sao Tome and Principe, Saudi Arabia, Senegal, Serbia, Slovakia, Slovenia, Somalia, South Africa, Spain, Sudan, Swaziland, Sweden, Switzerland, Syria, Tanzania, Togo, Tunisia, Turkey, Uganda, Ukraine, United Arab Emirates, United Kingdom, Vatican City, Western Sahara, Yemen, Zambia, Zimbabwe (Rhodesia)
eMLC - Enterprise MLC. Emerging class of MLC technology, uses wear leveling, bad-block mapping, improved error correction and write amplification techniques to improve the life of MLC flash storage. See MLC.
Emulex Corporation - http://www.emulex.com/
EOF - End of frame; a group of ordered sets used to delineate the end of a frame. Contrast with SOF.
EOL - End of Life; Disconnection of a particular product or service. EOL, in programming, often refers to the End of Line character. See LTB.
EOY - End of Year; Closure of projects on a calendar or fiscal year boundary.
EPROM - Erasable Programmable Read Only Memory; memory that uses floating-gate transistors and can be erased by strong ultraviolet light. Without the erasing light the memory stores data without electrical power. EPROM does not need a battery or other power source to retain data.
erasure coding - A method of data protection in which data is broken into fragments that are expanded and encoded with a configurable number of redundant pieces of data and stored across different location, such as disks, storage nodes or geographical locations.
The goal of erasure coding is to enable data that becomes corrupted to be reconstructed by using information about the data that is stored elsewhere in the array or even in another location.
Erasure coding works by creating a mathematical function to describe a set of numbers so that they can be checked for accuracy and recovered if one is lost. Otherwise know as polynomial interpolation or oversampling, this is the key concept behind erasure coding methods that are implemented most often using Reed-Solomon codes.
ERD - Engineering Requirements Document
ERM - Enterprise Resource Planning, integrates internal and external management information across an entire organization
ESM - Embedded System Management: Dell management utility and log for PowerEdge servers. See OMSA.
ESX - Enterprise-level computer virtualization software product adding management and reliability services on top of base hardware virtualization product. See ESXi.
ESX - VMware ESX is VMware's enterprise server virtualization platform. The platform is available in two versions: ESX Server (older) and ESXi Server (newer hypervisor architecture). VMware ESX and ESXi can be deployed as part of the VMware infrastructure (vSphere or VMware View) to enable centralized management for data center applications and enterprise desktops. One main feature of the enterprise server virtualization platform is that it allows users to contain server sprawl by running software appllications in virtual machines on fewer physical servers. See VMware.
ESXi - Enterprise-level computer virtualization software product adding management and reliability services on top of base hardware virtualization product but without the ESX Service Console and has a smaller footprint than ESX. See ESX.
ETL - Extract, Transform, Load: Usually used in the context of an ETL Database. Extract is the process of reading data from a database. Transform is the process of converting the extracted data from its previous form into the form it needs to be in sothat it can be place into another database. Transformation occurs by using rules or lookup tables or by combining the data with other data. Load is the process of writing the data into the target database. The three database functions, extract, transform and load, are combined into one tool to pull data out of one database and place it into another database.
EU - European Union - Union of 28 member states located primarily in Europe with ties for a common monetary union (19 member states), trade agreements and other standards and common policies.
ExaGrid - http://www.exagrid.com - ExaGrid Systems, Inc. is a disk-based backup hardware company. They employ a byte-level data deduplication appliance and use compression to minimize the amount of data stored.
F_Port - Fabric Port; On a fibre channel switch, a port that supports an N_Port. Contrast with E_Port, F_Port, FL_Port, G_Port, NL_Port.
FA - Failure Analysis or Field Authorization
FAI - First Article Inspection; Systematic and thorough inspection of the first fully produced product from manufacturing to guarantee that it meets the required specifications.
FakeRAID - Software RAID. FakeRAID is usually associated with disks that are lacking hardware RAID controllers. Instead, they are simply multi-channel disk controllers combined with special BIOS configuration options and software drivers to assist the OS in performing RAID operations. This gives the appearance of a hardware RAID, because the RAID configuration is done using a BIOS setup screen, and the operating system can be booted from the RAID.
The most common reason for using fakeRAID is in a dual-boot environment, where both Linux and Windows must be able to read and write to the same RAID partitions. Multiboot configurations are common among user who need multiple operating systems available on the sam emachine. These people shouldn't have to add a separate hard drive just so they can boot Linux. FakeRAID allows these users to access partitions interchangeably from either Linux or Windows. Another reason for using fakeRAID is if you define a disk mirror and a hard drive crashes, you can down the system and replace the failed drive and rebuild the mirror from the BIOS without having to boot into the operating system. Linux software RAID is more robust and better supported and thus, recommended over fakeRAID if you do not need to dual boot with Windows.
FC-0 - Fibre Channel Layer 0; physical interface Also see FC-1, FC-2, FC-3 and FC-4.
FC-1 - Fibre Channel Layer 1; encoding and decoding See 8b/10b encoding. Also see FC-0, FC-2, FC-3 and FC-4.
FC-2 - Fibre Channel Layer 2; framing protocol Also see FC-0, FC-1, FC-3 and FC-4.
FC-3 - Fibre Channel Layer 3; common services (e.g. striping, hunt groups, and multicast) Also see FC-0, FC-1, FC-2 and FC-4.
FC-4 - Fibre Channel Layer 4; application interface. Network and channel protocols for this layer include: SCSI, IPI, HiPPI, IP, AAL5, FC-LE, SBCCS, and IEEE 802.2. Also see FC-0, FC-1, FC-2, and FC-3.
FC-AL - Fibre Channel Arbitrated Loop - A shared 100MBps fibre channel transport supporting up to 126 devices and 1 fabric attachment.
FC-LE - Fibre Channel Link Encapsulation - Standard which gives a common precedure for encapsulating IEEE 802.2 Logical Link Control Protocol Data Units over Fibre Channel. See http://www.t11.org.
Fixed - Fixed; scheduling algorithm Contrast with MRU and RR.
FL_Port - Fabric Loop Port; On a fibre channel switch, a port that supports FC-AL devices. Contrast with E_Port, F_Port, G_Port, N_Port, NL_Port.
FOSS - Free and Open Source Software
- high-speed network technology primarily used to connection computer data storage devices. Fiber channel (American English spelling) refers to only optical cabling. Fibre channel (British English spelling) refers to either copper or optical fiber cabling. See FC-0, FC-1, FC-2, FC-3 and FC-4.
FLOGI - Fabric Login; a process by which a node establishes a logical connection to a fabric switch.
fabric - One or more fibre channel switches in a networked topology.
fabric attachment - Connection to a fibre channel network.
fixed capacity - A fixed capacity disk drive is that is guaranteed to be at least the advertised capacity. Contrast this with variable capacity disk drives.
formatted capacity - The actual capacity available to store data in a mass storage device. The formatted capacity is the gross capacity minus the capacity taken up by the overhead data required for formatting the media.
frame - A data unit comprising a start-of-frame delimiter, header, data payload, CRC, and end-of-frame delimiter.
FRRP - Dell Force10 Resilient Ring Protocol; Used on managed switches like Dell's Force10 S55.
FRU - Field Replaceable Unit; A system component expected to be replaced by a trained customer service engineer. See CRU
full duplex - Concurrent transmission and reception of data on a link.
FS - File System
FTOS - Force10 Operating System; FTOS is the firmware family running on Force10 Ethernet switches such as the Dell Force10 S55. FTOS runs on NetBSD.
G_Port - Generic Port; On a fibre channel switch, a port that supports either F_Port or E_Port functionality. Contrast with E_Port, F_Port, FL_Port, N_Port, NL_Port.
GA - General Availability; Designation of a product or service that has been officially released for wide distribution.
GaaS - Gaming-as-a-Service Cloud Service Model; The NVIDIA GeForce Grid
Two major advantages or benefits of NVIDIA GeForce Grid is the ability to minimize latency and lag, and the ability to maximize game server scaling and efficiency.
With NVIDIA GeForce GRID technology, gamers will have the freedom to play games from the cloud on any display, including TVs, monitors, notebooks, tablets, and even smartphones.
First-generation cloud gaming platforms suffer from long latency or lag, low-quality graphics and high costs because of a 1:1 ratio of game streams to data center servers. NVIDIA GeForce GRID technology resolves these obstacles and will make game streaming as common as renting a movie online.
The NVIDIA Kepler Cloud GPU
Jen-Hsun Huang, NVIDIA president and CEO believed that “Kepler cloud GPU technologies shifts cloud computing into a new gear.The GPU has become indispensable. It is central to the experience of gamers. It is vital to digital artists realizing their imagination. It is essential for touch devices to deliver silky smooth and beautiful graphics. And now, the cloud GPU will deliver amazing experiences to those who work remotely and gamers looking to play untethered from a PC or console.”
The NVIDIA Tesla K10 GPU and Tesla K20 GPU are two computing accelerators that are built and designed to handle the most complicated high performance computing problems. The NVIDIA Tesla K10 and Tesla K20 are three times better that Fermi based GPUs.
The NVIDIA VGX Platform
From NVIDIA’s press release: NVIDIA unveiled the NVIDIA VGX platform, which enables IT departments to deliver a virtualized desktop with the graphics and GPU computing performance of a PC or workstation to employees using any connected device.
With the NVIDIA VGX platform in the data center, employees can now access a true cloud PC from any device — thin client, laptop, tablet or smartphone — regardless of its operating system, and enjoy a responsive experience for the full spectrum of applications previously only available on an office PC.
NVIDIA VGX enables knowledge workers for the first time to access a GPU-accelerated desktop similar to a traditional local PC. The platform’s manageability options and ultra-low latency remote display capabilities extend this convenience to those using 3D design and simulation tools, which had previously been too intensive for a virtualized desktop.
Integrating the VGX platform into the corporate network also enables enterprise IT departments to address the complex challenges of “BYOD” — employees bringing their own computing device to work. It delivers a remote desktop to these devices, providing users the same access they have on their desktop terminal. At the same time, it helps reduce overall IT spend, improve data security and minimize data center complexity.
"NVIDIA VGX represents a new era in desktop virtualization," said Jeff Brown, general manager of the Professional Solutions Group at NVIDIA. “It delivers an experience nearly indistinguishable from a full desktop while substantially lowering the cost of a virtualized PC.”
The NVIDIA VGX platform is part of a series of announcements NVIDIA is making today at the GPU Technology Conference (GTC), all of which can be accessed in the GTC online press room.
The VGX platform addresses key challenges faced by global enterprises, which are under constant pressure both to control operating costs and to use IT as a competitive edge that allows their workforces to achieve greater productivity and deliver new products faster. Delivering virtualized desktops can also minimize the security risks inherent in sharing critical data and intellectual property with an increasingly internationalized workforce.
NVIDIA VGX is based on three key technology breakthroughs:
NVIDIA VGX Boards. These are designed for hosting large numbers of users in an energy-efficient way. The first NVIDA VGX board is configured with four GPUs and 16 GB of memory, and fits into the industry-standard PCI Express interface in servers.
NVIDIA VGX GPU Hypervisor. This software layer integrates into commercial hypervisors, such as the Citrix XenServer, enabling virtualization of the GPU.
NVIDIA User Selectable Machines (USMs). This manageability option allows enterprises to configure the graphics capabilities delivered to individual users in the network, based on their demands. Capabilities range from true PC experiences available with the NVIDIA standard USM to enhanced professional 3D design and engineering experiences with NVIDIA Quadro or NVIDIA NVS GPUs.
The NVIDIA VGX platform enables up to 100 users to be served from a single server powered by one VGX board, dramatically improving user density on a single server compared with traditional virtual desktop infrastructure (VDI) solutions. It sharply reduces such issues as latency, sluggish interaction and limited application support, all of which are associated with traditional VDI solutions.
With the NVIDIA VGX platform, IT departments can serve every user in the organization — from knowledge workers to designers —with true PC-like interactive desktops and applications.
NVIDIA VGX Boards
NVIDIA VGX boards are the world’s first GPU boards designed for data centers. The initial NVIDIA VGX board features four GPUs, each with 192 NVIDIA CUDA® architecture cores and 4 GB of frame buffer. Designed to be passively cooled, the board fits within existing server-based platforms.
The boards benefit from a range of advancements, including hardware virtualization, which enables many users who are running hosted virtual desktops to share a single GPU and enjoy a rich, interactive graphics experience; support for low-latency remote display, which greatly reduces the lag currently experienced by users; and, redesigned shader technology to deliver higher power efficiency.
NVIDIA VGX GPU Hypervisor
The NVIDIA VGX GPU Hypervisor is a software layer that integrates into a commercial hypervisor, enabling access to virtualized GPU resources. This allows multiple users to share common hardware and ensure virtual machines running on a single server have protected access to critical resources. As a result, a single server can now economically support a higher density of users, while providing native graphics and GPU computing performance.
This new technology is being integrated by leading virtualization companies, such as Citrix, to add full hardware graphics acceleration to their full range of VDI products. l
NVIDIA User Selectable Machines
NVIDIA User Selectable Machines or USMs allow the NVIDIA VGX platform to deliver the advanced experience of professional GPUs to those requiring them across an enterprise. This enables IT departments to easily support multiple types of users from a single server.
USMs allow better utilization of hardware resources, with the flexibility to configure and deploy new users’ desktops based on changing enterprise needs. This is particularly valuable for companies providing infrastructure as a service, as they can repurpose GPU-accelerated servers to meet changing demand throughout the day, week or season.
Compare IaaS, PaaS, SaaS, HaaS, and GaaS.
GBIC - Gigabit Interface Converter; a removable transceiver module for fibre channel and gigabit ethernet physical-layer transport. GBICs are a critical component of the Fibre Channel transport and are the most commonly used transceivers for Fibre Channel applications. First generation optical GBICs used edge-emitting lasers common in CD players which, for high-speed data transport, were expensive to manufacture, test, and consumed more power, radiated more heat, and were more susceptible to loss of calibration over time than modern VCSELs.
Gbps - Gigabits per second.
GB - 10^9 = (1000 x 1000 x 1000) bytes. This is the IEC standard (although not universally accepted). Contrast this with GiB.
GiB - about 1.074 x 10^9 = (1024 x 1024 x 1024) bytes. This is the IEC standard (although not widely accepted). Contrast this with GB.
Gigabit - For fibre channel, 1,062,500,000 bits per second.
GLM - Gigabit link module; a semipermanent transceiver that incorporates serializing/deserializing functions.
Google Compute Engine - A type of IaaS where users can use the very same infrastructure/system the Google is using.
Google App Engine - These service is for developers where they can create and run their applications.
Google Apps - Google Apps offers different productivity service for user.
GOST - Russian technical standards for products maintained by the Euro-Asian Council for Standardization, Metrology and Certification (EASC). GOST certification is required to sell products in Russia, Belarus, Ukraine, Moldova, Kazakhstan, Azerbaijan, Armenia, Kyrgyzstan, Uzbekistan, Tajikistan, Georgia and Turkmenistan.
GTM - Go To Market
HA - High Availability; Device with multiple backup power supplies, disk drives, and/or standby devices with the idea that the function provided by the device will always be available in spite of failures.
HA-SMR - Host Aware SMR disk drive
HaaS - Hardware-as-a-service Cloud Service Model;
Compare IaaS, PaaS, SaaS, HaaS, and GaaS.
Hadoop - Open-source software framework for storing and processing big data in a distributed fashion on large clusters of commodity hardware. Essentially, Hadoop accomplishes two tasks: massive data storage and faster processing.
Hadoop is named after the main developer's, Doug Cutting, son's toy elephant.
Key elements of Hadoop:
The Apache Hadoop project is a major source of Hadoop.
Hadoop Bestiary:
HGST - Formerly Hitachi Global Storage Technologies; Now a WD company. See http://www.hgst.com.
HAL - Hardware Abstraction Layer (Amplidata's Python scripts manipulate)
HAMR - Heat Assisted Magnetic Recording with TDMR and/or SMR
HBA - Host bus adapter; an interface between a server or workstation bus and the fibre channel network.
HDD - Hard Disk Drive (HDD) or Hard Drive (HD); a hardware disk used as a computer's main storage media device to permanently store data. Connections are often made using a variant of either ATA, SCSI, or SATA.
HDMR - Heated-Dot Magnetic Recording (BPMR + HAMR)
HDMI - High Definition Multimedia Interface - Industry standard for connecting high-definition audio and video devices.
HDS - http://hds.com/ - Hitachi Data Systems (HDS); Large provider of data storage systems. Major competitors in data storage systems are EMC, HP, IBM, and NetApp.
HiPPI - High-Performance Parallel Interface; a high-speed interface normally used in supercomputer environments.
HM-SMR - Host Managed SMR disk drive
Compare with HA-SMR
hot-plug - Hot pluggable devices can be disconnected/connected to their power supplies while running without damage.
HP - Hewlett-Packard Company (HP); Large provider of data storage systems. Major competitors in data storage systems are EMC, HDS. IBM and NetApp.
HSSDC - High-speed serial direct connect; a form factor that allows quick connect/disconnect for copper interfaces.
HTTP - HyperText Transfer Protocol; protocol commonly used for text markup and data transfer
hub - In fibre channel, a wiring concentrator that collapses a loop topology into a physical star topology.
Hybrid Cloud - This is somewhere in between Private Cloud and Public Cloud.
hyperconverged - Marketing term describing the embedding of virtualization hardware in with the rest of a system the purpose being to provide a fully scalable and extendable configuration.
IaaS - Infrastructure-as-a-Service Cloud Service Model; In this type of service model, virtual machines, servers, storage devices and other physical materials are considered as assets. People who are looking for firewalls, load balancers, physical computers should opt for this model.
Usually start up companies who are still testing their systems uses this kind of model. Cloud users installs their operating system and/or application software in the Cloud Service Provider's machines. But it’s the responsibility of the Cloud user to maintain the OS and application they install.
In laaS model, the service is charged like a utility bill for using their machines. The Cloud Providers charge the Cloud users like how many GB of storage was used, or how much network bandwidth was consumed or its CPU usage and the likes.
An example of IaaS is AWS. Amazon Cloud Computing provides one of the cost effective ways for your project or business.
Advantages of Infrastructure as a Service (laaS) are that capital expenditure on hardware can be avoided. You do not have to buy computers, hardware and stuff. It can easily be entered because of low barriers and scaling is done automatically.
Disadvantages of Infrastructure as a Service (laaS) is that the Cloud User is dependent on the service that the Cloud Providers provide. Efficiency and effectiveness of the Cloud User or its business can not go beyond what the Cloud service provider has to offer.
Compare IaaS, PaaS, SaaS, HaaS, and GaaS.
IBM - http://ibm.com/ - IBM Corporation (International Business Machine); Large provider of data storage systems. Major competitors in data storage systems are EMC, HDS, HP. and NetApp.
IB Trackable - Install Base Trackable; IB Trackable typically refers to hardware (e.g. disks) that do not report or identify themselves to services such as Oracle's Service Fulfullment Manager (SFM).
Icinga - Open source computer system and network monitoring application. See Nagios.
IDNF - Identifier Not Found: WD disk drive error. The drive positions the heads on track via servo and expects to read the ID of the track to start the data transfer. The ID of the track is not found and an IDNF error is reported to the host. No data is transferred.
IDO2DB - Icinga Data Out to Database: Service daemon that uses SSL encrypted TCP sockets to send Icinga data to a database (e.g. MySQL, PostgreSQL, MariaDB).
iDRAC - Integrated Dell Remote Access Controller; Dell's frontage controller PC for their servers;
IEC - International Electrotechnical Commission - http://www.iec.ch
IEEE - Institute of Electrical and Electronics Engineers - http://www.ieee.org
IGMP - Internet Group Management Protocol; Used on managed switches like Dell's Force10 S55.
Imation Corp. - http://www.imation.com/en-US/ Imation originally started as the data storage division of 3M in the 1950s.
in-band - Transmission of management protocol over the fibre channel transport.
INCITS - InterNational Committee for Information Technology Standards - Organization devoted to creating and disseminating technology standards. See http://www.incits.org. Also see SCSI Storage Interfaces: T10 Fibre Channel Interfaces: T11 and ATA Storage Interfaces: T13.
InfiniBand - IB - Commmunications link used in high-performance computing featuring very high throughput and very low latency. It is designed to be scalable and uses a switched fabric network topology.
Infiniband competes with gigabit ethernet. As of 2014, Mellanox InfiniBand switches offer 40 to 56 Gb/s per port.
INRIA - French Institute for Research in Computer Science and Automation (Institut National de Recherche en Informatique et en Automatique): Known for such projects as Caml, XtreemFS (object-based distributed file system) and Bigloo (Scheme implementation).
initiator - On a fibre channel network, typically a server or a workstation that initiates transactions to disk or tape targets.
inform - A SNMP inform is an unsolicited message sent to a notification receiver. The inform was not introduced into the protocol until SNMP version 2c, so it is not available in devices that only support SNMP version 1. Unlike SNMP traps, informs are confirmed messages, and thus are more appropriate for critical notifications.
Intel - Intel Corporation; http://www.intel.com - Compute, memory and peripheral semiconductor chip developer and manufacturer. Known for their PC computer chips.
I/O - Input/Output (I/O); data stream in either direction from/to a device
IOPS - Disk Input/Output Operations Per Second; Common performance measurement used to benchmark computer storage devices. iometer is an Intel developed program commonly used to measure IOPS. For hard disk drives, the random IOPS numbers are primarily dependent upon seek time. For solid state storage, the random IOPS number are primarily dependent upon the devices internal controller and memory interface speeds.
IP - Internet Protocol
IPI - Intelligent Peripheral Interface
IPMI - Intelligent Platform Management Interface; System administrator interface used to manage and monitor a computer system. See BMC.
intercabinet - A specification for copper cabling that allows up to 30 meter distance between enclosures.
IR - Integrated RAID mode; Disk controller firmware that uses RAID smarts. Compare with IT
iSCSI - Internet Small Computer Systems Interface; Internet SCSI
See SCSI
IT - Initiator Target mode; Disk controller firmware that just acts as a host bus adapter without trying to do anything smart. Compare with IR
ISO - International Organization for Standardization; An ISO image or .iso file is an archive file that contains a disk image in ISO 9660 file system format.
ISP - Internet Service Provider
ISV - Independent Software Vendor
iWARP - Internet Wide Area RDMA Protocol
JBOD - Just a Bunch of Disks; Typically configured as a fibre channel FC-AL segment in a single chassis. Contrast with ABOD, EBOD, RBOD, SBOD, and JBOK.
JBOF - Just a Bunch of Flash; SanDisk's term for their InfiniFlash storage system
JBOK - Just a Bunch of Kinetics; Contrast with ABOD, EBOD, RBOD, SBOD. See Kinetic Hard Drive.
jitter - Deviation in timing that a bit stream encounters as it traverses a physical medium.
JSON - JavaScript Object Notation
K28.5 - A special 10-bit character used to indicate the beginning of a fibre channel command.
KHD - Kinetic Hard Drive - Seagate's modified 4TB Terascale drive. The terascale is a nearline 3.5in disk spinning at 5,900 RPM. It swaps the SATA or SAS interface connections for two 1Gbps SGMII Ethernet ports to directly attach to the network. The drives repurpose the standard SAS HDD connector.
Seagate Kinetic Drive (as of 2014)
KB - 1000 bytes. This is the IEC standard (although not universally accepted). Contrast this with KiB.
KiB - 1024 bytes. This is the IEC standard (although not universally accepted). Contrast this with KB.
KPI - Key Performance Indicator
Kryder's Law - Disk drive bit density doubles every two years. Named after Mark Kryder.
Current drives are quite close to the 1Tbit/in2 superparamagnetic limit of magnetic media. See SMR.
See Moore's Law.
KVM - Keyboard, Video and Mouse; A KVM switch is a hardware device that allows a user to control multiple computer from one or more sets of keybaords, video monitors and mouse.
See KVM.
KVM - Kernel-based Virtual Machine; A virtualization for the Linux kernel that turns it into a hypervisor.
See KVM.
LACP - Link Aggregation Control Protocol; IEEE 802.3ad standard that provides a way for two systems to automatically establish and maintain LAGs. LACP bonding refers to any of a variety of methods to combine multiple network connections in parallel in order to increase throughput beyond what a single connection could sustain and to provide redundancy in case one of the links should fail.
LAG - Link Aggregation Group; Link aggregation is a method of grouping multiple physical interfaces into a single logical interface. A LAG is a port channel. A LAG is a group of links that appear to a MAC client as if they were a single link. In FTOS, a LAG is referred to as a port channel interface.
LAN - Local area network; a network linking multiple devices in a single geographical location
LASER - Light Amplification by Stimulated Emission of Radiation: A device that emits coherent light via optical amplifaction. See GBIC.
latency - the period of time that the read/write heads of a disk drive wait for the disk to rotate the data to an accessible position. For a disk rotating at 10,000 RPM, the average latency is 3 milliseconds.
LBA - Logical Block Address: An alternative addressing methodology of identifying a given location on a SATA drive that permits disk sizes greater than 528 MB.
Lattus - Object storage system sold by Quantum. Internally uses Arakoon and OCaml
LDPC - Low-Density Parity Check: A linear error correcting code.
LED - Light-Emitting Diode; For fibre channel connections, typically a status indicator on an interconnect device.
LIP - Loop Initialization Primitive; used to initiate a procedure that results in unique addressing for all nodes, to indicate a loop failure, or to reset a specific node.
LISM - Loop Initialization Select Master; a process by which a temporary loop master is determined.
LLDP - Link Layer Detection Protocol; Used on managed switches like Dell's Force10 S55.
load balancing - A method by which multiple paths from/to a data array are used optimally.
loop port state machine - Logic that monitors and performs the tasks required for initialization and access to the loop.
LTFS - Linear Tape File System: Part of the SNIA family of standards in 2012.
LTO - Linear Tape-Open: Magnetic tape data storage technology standard developed by the LTO Consortium (Hewlett-Packard, IBM and Seagate).
LUN - Logical Unit Number; a logical device that is identified by a locally unique number.
LTB - Last Time Buy: The last order that can be made of a product or service that is to be discontinued. See EOL.
LTS - Long Term Support: Often used in conjunction with a particular operating system release (e.g. Ubuntu 14.04.2 LTS). unique number.
LOM - Lights out Management: Used on servers such as HP (iLO) or Dell to provide connectivity without the use of a dedicated NIC. A LOM is an out-of-band management port.
MAC address - Media Access Control address, globally unique identifier assigned to a network interface controller (NIC). This is useful to help identify every hardware device on a network. The identifier is typically is written as six two digit hexadecimal numbers separated by colons (e.g. A4:B3:4F:23:FA:CC).
maturity - Extensive long term test of product before release.
MB - 1000000 = (1000 x 1000) bytes. This is the IEC standard (although not universally accepted). Contrast this with MiB.
Mbps - Megabits per second.
MBps - Megabytes per second.
MCE - Machine Check Exception: A type of computer hardware error that occurs when a computer's central processing unit detects a hardware problem. On Linux, when a MCE is detected, a process (such as mcelog) will write a message to the kernel log and/or the console screen.
mcelog - Machine Check Exception Log: A log of MCE events. The Linux kernel for x86 CPUs no longer decodes and logs recoverable machine check exception (MCE) events to the kernel log on tis own. Instead, the MCE data is kept in a buffer which can be read from userspace via the /dev/mcelog device node. The mcelog tool is used to collect and decode MCE events. It will log the decoded MCE events to syslog.
MCN - Lattus Management Controller Node; (e.g. C5, C10)
MD - metadata
MDADM - Linux utility used to manage software RAID devices. Replacement for Linux DMRAID command for software raid manipulation. See DMRAID.
MDC - MetaData Controller
Mellanox - Mellanox Technologies http://www.mellanox.com/
Leader in Infiniband technology.
MEM - Management Extension Modules; PSA specific vendor plug-ins.
MEP - Maintenance End Point; Points at the edge of the domain that define the boundary and confine CFM messages within the boundary. MEPs are inward facing by default. Inward facing means that they communicate through the relay function side, not the wire side (connected to the port), whereas MEPs that can be configured as outward facing communicate through the wire side, and not through the relay function side. Used on managed switches like Dell's Force10 S55.
metadata - Data describing a collection of data.
MetaStore - Lattus system of processes that store metadata of objects, superblocks, spreads, policies and namespaces. The underlying technology is the Arakoon distributed key-value store.
MiB - 1048576 = (1024 x 1024) bytes. This is the IEC standard (although not universally accepted). Contrast this with MB.
MIA - Media Interface Adapter; a device that converts optical signaling to electrical.
MIB - Management Information Base; an SNMP structure for device management.
MicroFemto Slider - Disk drives for which the read/write head is mounted on a small, lightweight microfemto slider that allows the head to move more quickly from track to track on the disk.
MIME - Multipurpose Internet Mail Extensions; Protocol used to send extended mail contents (e.g. foreign languages, file attachments, pictures).
MLC - Multi-Level Cell: Type of solid-state device where an individual cell of memory is used to store more than one bit of information. MLC NAND flash refers to flash memory. Constrast MLC with SLC. See MLC.
MOFCOM - Ministry of Commerce of China; regulatory body for technology in China
Moore's Law - Transistor density doubles every two years. Named after Gordon E. Moore.
See Kryder's Law.
MountainGate - MountainGate Imaging Systems Corporation; Acquired by ADIC in September 1999. See Quantum.
MPIO - Multi-Path Input Output
MPM - Multipathing MEM; Prior to vSphere 4.1, this was called Multipathing Plugins (MPP) Also see VMware.
MPP - Multipathing Plugins; Starting with vSphere 4.1, the new name is Multipathing MEM (MPM). Also see VMware.
MRD - Marketing Requirements Document
MRU - Most Recently Used; scheduling algorithm Contrast with MRU and Fixed.
MS - Microsoft, Corporation (MS)
MSTP - Multiple Spanning Tree Protocol; Used on managed switches like Dell's Force10 S55.
MTBF - Mean Time Before Failure; Arithmetic mean (a.k.a. average) time before failure
MTU - Maximum Transmission Unit; For a network switch, this is the maximum allowed size of a transmission unit (e.g. packet). The MTU is usually set during switch configuration. For instance, Jumbo Frames on a Dell S55 Gigabit Ethernet port would have a MTU of 9252.
multimode - A fiber-optic cabling specification that allows up to 500 meter distance between devices. Multimode fiber has a large diametral core that allows multiple modes of light ot propagate. Because of this, the number of light reflections created as the light passes through the core increases, creating the ability for more data to pass through at a given time. Because of the high dispersion and attenuation rate with this type of fiber, the quality of the signal is reduced over long distances. Multimode fiber is typically used for short distance, data and audio/video applications in LANs, RF broadband signalds, such as what cable companies commonly use, cannot be transmitted over multimode fiber. Contrast with single mode
multipath - Multipath I/O is a fault-tolerance and performance enhancement technique for computer storage involving multiple physical paths between the server and its mass storage devices. The paths can involve buses, controllers, switches, hubs, and other bridge devices.
N_Port - Node Port; A fibre channel port in a point-to-point or fabric connection. Contrast with E_Port, F_Port, FL_Port, G_Port, NL_Port.
NAS - Network-Attached Storage; a disk array connected to a controller that provides connection to a LAN transport. NAS is block based at the level of the disks within a NAS array, but uses a file interface in between the application and the disks. Contrast with OSD.
NIC - Network Interface Controller, a computer hardware component that connects a computer to a computer network. A NIC is also know as a Network Interface Card.
NCQ - Native Command Queuing: Improvement of disk drive performance through intelligent re-ordering of the I/O requests so they read/write to and from the nearest available sectors and minimize the need for additional disk revolutions or head actuator movement.
NDMP - Network Data Management Protocol; a protocol for performing tape backups without consuming server resource.
NEBS - Network Equipment-Building System - Telecommunications equipment set of standards in the United States. An interesting impact for computer systems is the requirement for a 48 volt DC power system.
Nexenta - Company specializing in software-defined storage (OpenSDS). See http://www.nexenta.com
NDC - Network Daughter Card See rNDC.
NDP - Neighbor Discovery Protocol; Used on managed switches like Dell's Force10 S55.
NetApp - http://www.netapp.com/ - NetApp, Inc.; Specializes in computer storage and data management products. Major competitors in data storage systems are EMC, HDS. HP and IBM.
Nexsan - Products include Assureon Secure Storage, E-Series High Density Storage, and NST Hybrid Storage. See Imation Corp. Corp.
NFS - Network File System
Nimble Storage Inc. - http://www.nimblestorage.com/
NL_Port - Node Loop Port; a port that supports FC-AL protocol Contrast with E_Port, F_Port, FL_Port, G_Port, N_Port, NL_Port.
NNTP - Network News Transfer Protocol: News transfer protocol governed by RFC 977.
NMP - Native MultiPathing Also see VMware.
node - A bibre channel entity that supports one or more ports.
node_name - A unique 64-bit identifier assigned to a fibre channel node.
non-OFC - Laser transceivers whose lower-intensity output does not require special OFC mechanisms.
notification - A SNMP notification is an unsolicited message sent to a notification receiver (a.k.a. Network Management System (NMS) or Element Management System (EMS). Notifications generally indicate a status change on a device. It could be information about the change, or a message to indicate an alarm condition. SNMP defines two types of notifications: traps and informs.
NRPE - Nagios Remote Plugin Executor; Client side agent to support Nagios and other monitoring utilities (e.g. Icinga).
NTP - Network Time Protocol
NVM - Non-volatile Memory; memory that retains data after the host device's power is turned off.
NVMe - Non-volatile Memory Express; NVMe is a standard based on PCIe and built for physical slot architecture. NVMe most commonly refers to flash memory.
NVMe provides much greater bandwidth than SAS and SATA, with vastly improved queuing. SATA III provides 6 Gbps and 600 Mbps throughput. SAS provides 12 Gbps and 8Gbps throughput. NVMe has data transfer performance characters of PCIe Gen 3 and therefore has bandwidth of around 1 Gbps per lane, up to 16 Gbps in a 16-lane configuration. PCIe Gen 4, expected in late 2017, is expected to double that.
NMVe is built to handle more queues than SAS and SATA at the drive, with 65K queues and 65K command queue depth. This compares to one queue for SAS and SATA and queue depths of 254 for SAS and 32 for SATA.
NVRAM - Non-volatile Random Access Memory; memory that retains data after the host device's power is turned off. Two common types of NVRAM are SRAM and EEPROM.
OAM - Ethernet Operations, Administration, and Maintenance is a protocol for installing, monitoring, and troubleshooting Ethernet networks to enhance management within the Ethernet infrastructure. This is defined by IEEE 802.3ah. Used on managed switches like Dell's Force10 S55.
OBE - Out of Box Experience Report; Test installation and setup of a product straight out of the shipping boxes in simulation of a customer's experience in order to find any problem before product shipment to an actual customer.
object store - A repository of objects where the definition of object is a full or partial collection of associated data along with data describing the data, the metadata.
OCAML - Object-oriented Categorical Abstract Machine Language. See Caml
OEM - Original Equipment Manufacturer
OFC - Open Fiber Control; a method used to disable and enable laser signaling for higher-intensity laser transceivers.
OID - Object Identifier: An SNMP OID is a series of integers that uniquely identify a data object. OIDs can be represented in two ways. The first is using raw numbers (e.g. .1.3.6.1.2.1.1.5.0) or the second is using names to represent the numbers (e.g. .iso.org.dod.internet.mgmt.mib-2.system.sysName.0). Raw number and names can be freely mixed in the same OID (e.g. .iso.org.6.1.mgmt.1.1.sysName.0).
OLTP - Online Transaction Processing
OMSA - OpenManage Server Administrator: Dell web based application to manage Dell PowerEdge Servers. This computer utility can perform proactive system monitoring, system diagnostics, troubleshoot hardware issues, configure RAID devices, view and manage ESM logs and more.
Oracle - Oracle Corporation is a computer technology company specializing in computer hardware systems and enterprise software products. See http://www.oracle.com.
OpenManage - OpenManage is a hardware management application from Dell Corporation It provides information and control capabilities of systems, devices and components. See http://www.dell.com.
ordered set - A group of low-level protocols used to manage frame transport, initialization, and media access.
OSD - Object-based Storage Device; Storage objects are variable size logical collections of bytes. The collections are identified to an application and accessed using metadata. Contrast with NAS.
OSMI - Object Store Management Interface; Python script for various Lattus maintenance and configuration operations. The script was written by and is maintained by Amplidata. The code can be found on the Lattus controller node at /opt/qbase3/apps/osmi/osmi
OSPF - Open Shortest Path First; Used on managed switches like Dell's Force10 S55.
out-of-band - Transmission of management protocol outside of the fibre channel network, typically over Ethernet.
Overland Storage Inc. - http://www.overlandstorage.com/
PaaS - Platform-as-a-Service Cloud Service Model; In this type of service model, this involves database, web server, operating systems, programming languages and the likes. Cloud service providers provides a computing platform where application developers can test and run their developed software through the Cloud Service provider’s platform, without the need to spend for additional hardware to test their software.
Generally this type of Cloud Computing Service model is for application or software developers, testers, administrators and the likes. This type of service is typically used to developed SaaS (Software as a Service) applications.
Compare IaaS, PaaS, SaaS, HaaS, and GaaS.
parallel - The simultaneous transmission of multiple data bits over multiple lines.
Pay as you go - A type of subscription or payment option where you just need to pay for the resources that you have used.
passive copper - A low-cost fibre channel connection that allows up to 13 meter copper cable lengths.
PCI - Peripheral Component Interconnect: PCI is a local bus standard developed by Intel Corporation. Most modern PCs include a PCI bus in addition to a more general ISA expansion bus. PCI is also used on some versions of the Macintosh computer. PCI is a 64-bit bus, though it is usually implemented as a 32-bit bus. It can run at clock speeds of 33 or 66 MHz. At 32 bits and 33 MHz, it yields a throughput rate of 133 MBps. Although it was developed by Intel, PCI is not tied to any particualr family of microprocessors.
PCIe - Peripheral Component Interconnect Express
PCM - Phase-change Memory; New memory technology being developed by IBM.
PCM materials exist in either in amorphous or crystalline state, with each state exhibiting different levels of electrical conductivity. Data is stored in binary, so one level can be used to represent a 0 and the other to represent a 1 on each PCM cell, with a low voltage then applied to read the data bit back.
This new technology promises to write and retrieve data 100 times faster than flash memory. Also, PCM promises to last 10 million write cycles compared to the 3,000 of a standard flash-based USB stick. PCM does not lose data when switched off like DRAM. Currently the major problems with PCM include low density and therefore cost.
PDU - Protocol Data Unit: Information that is to be delivered as a bundled unit. This is simply a data packet. See SNMP.
PDU - Power Distribution Unit (essentially a long power strip) See FC-LE.
PERC - PowerEdge© Expandable RAID Controller: Dell's line of RAID controller cards for their PowerEdge servers. The card typically contains a SAS host interface with high speed cache support. The card also supplies support for and JBOD configurations. Note that the SAS controller can run both SAS and SATA drives.
PI - PIM-Spare Mode; Used on managed switches like Dell's Force10 S55.
PI - PIM-Source Specific Mode; Used on managed switches like Dell's Force10 S55.
Pivot3 - Pivot3: Surveillance storage company founded in 2003. See http://www.pivot3.com.
PLOGI - A port-to-port login process by which initialtors establish sessions with targets.
PMR - Perpendicular Magnetic Recording; The magnetization of each data bit is aligned vertically to the spinning idsk, rather than longitudinally as has been the case in hard dirve technology for decades. This enables more data on a given disk than is possible with conventional longitudinal recording, and provides a platform for future expandion of hard drive densities. Sometimes PMR refers to PMR with TDMR and/or SMR.
point-to-point - A dedicated fibre channel connection between two devices.
port - A fibre channel physical entity that connects a node to the network.
port aggregation - See port channel
port channel - Port Channels are a quick way to get more bandwidth by aggregating multiple connections in one virtual pipe. For instance, rie four gigabit ports together into a channel and it becomes a four gigabit connection. Channels can also offer redundancy fault tolerance for physical connections. If one of the links involved in a channel loses connection, the channel will continue on with the existing ports and three quarters of the bandwidth. For the Lattus system, port channels are only used to cascade multiple switches.
port_name - A unique 64-bit identifier assigned to a fibre channel port.
Postgres - PostgreSQL or Postgres is a cross-platform object-relational database management system that implements most of the SQL:2011 standard. Postgres comes from the Ingres project at the Univeristy of California, Berkeley. The name comes from its origins as a "post-ingres" database developed from the University Ingres DBMS (INteractive Graphics Retrieval System Database Management System).
power factor correction - Power factor correction is sometimes applied to a "dirty" inductive load, where the electrical effect of the load causes the phase angle to change which means the power used is less effective, so relatively costs more to run. Correction addresses this. Smart power supplies often have a setting to turn on and off correction.
PRD - Product Requirements Document
primitive sequences - Ordered sets that indicate or initiate state changes on the transport and require at least three consecutive occurrences to trigger a response.
primitive signals - Ordered sets that indicate actions or events and require only one occurrence to trigger a response.
Private Cloud - A network of cloud computing where the infrastructure is owned and used by a private company
private loop - A free-standing FC-AL with no fabric attachment.
private loop device - An FC-AL device that supports fabric login and services.
Progressive Capacitytm - WDA> Progressive Capacity disk drives are WD Ae drives for which a carton of drives (20 count) has a mean storage capacity. Some of the drives may have less than the mean capacity and some more.
PSA - Pluggable Storage Architecture; special layer of the monitor (VMkernel) used to manage storage multipathing. VMware PSA allows storage vendors, through a modular storage architecture (a.k.a. PSA), to write plug-ins for their specific capabilities. Also see VMware.
PSM - Path Selection MEM; Prior to vSphere 4.1, this was called Path Selection Plugin (PSP). Also see VMware.
PSP - Path Selection Plugin; Starting with vSphere 4.1, the new name is Path Selection MEM (PSM). Also see VMware.
PSU - Power Supply Unit
PTR - Problem Tracking Report (a.k.a. bug); For engineering failure analysis and corrective action.
Public Cloud - A network of cloud computing where the infrastructure or service is open to the public for their own use.
public loop - An FC-AL attached to a fabric switch.
public loop device - An FC-AL device that supports fabric login and services.
PV - Paravirtualized; A virtualization technique that presents a software interface to virtual machines that is similar, but not identical to that of the underlying hardware. The virtual host must be aware of and make use of special hooks provided. The intent is to speed execution by reducing the overall performance degradation of machine-execution inside the virtual-guest.
PVLAN - Private VLAN; Used on managed switches like Dell's Force10 S55.
PVST+ - Per-VLAN Spanning Tree plus; Used on managed switches like Dell's Force10 S55.
PXE - Preboot Execution Environment;
PWLtm - Pre-emptive Wear Leveling: This WDA> feature provides a solution for protecting the recording media of a disk drive against mechanical wear. In cases where the dirve is so busy with incoming commands that it is forced to stay in a same cylinder position for a long time, the PWL control engine initiates forced seeks so that disk lubricant maintins an even distribution and does not become depleted. This feature ensures reliability for applications that perform a high incidence of read/write operations at the same physical location on the disk.
PyMonkey - Python module that offers a low-level interface to the Mozilla SpiderMonkey JavaScript engine.
PPZT - Piezo Electric Transducer
QoS - Quality of Service
Quantum - Quantum Corporation; http://www.quantum.com - Provider of scale-out storage, archive and data protection products. Specialists in data deduplication backup and recovery, physical, virtual and cloud environments.
queue depth - A port can only service one request at a time and additional requests are placed in the port queue to be serviced when the current request has been completed. Once the port queue depth is reached, any further requests are rejected until space in the queue becomes available.
See NMVe.
Remote Access Controller Admintm - Dell command line tool that allows remote or local management of Dell servers via the iDRAC or DRAC.
RAFFtm - Rotary Acceleration Feed Forward: WD drives that employ technology to maintain hard drive performance in high vibration environments through adaptive compensation of the servo system.
RAID - Redundant Array of Independent Disks
RAID 0 - RAID Level 0: Striped Disk Array without Fault Tolerance: Provides data striping (spreading out blocks of each file across multiple disk drives) but no redundancy. This improves performance but does not deliver fault tolerance. If one drive fails then all data in the array is lost. See RAID 1, RAID 2, RAID 3, RAID 4, RAID 5, and RAID 6.
RAID 1 - RAID Level 1: Mirroring and Duplexing: Provides disk mirroring. Level 1 provides twice the read transaction rate of single disks and the same write transaction rate as single disks. See RAID 0, RAID 2, RAID 3, RAID 4, RAID 5, and RAID 6.
RAID 2 - RAID Level 2: Error-Correcting Coding: Not a typical implementation and rarely used, Level 2 stripes data at the bit level rather than the block level. See RAID 0, RAID 1, RAID 3, RAID 4, RAID 5, and RAID 6.
RAID 3 - RAID Level 3: Bit-Interleaved Parity: Provides byte-level striping with a dedicated parity disk. Level 3, which cannot service simultaneous multiple requests, also is rarely used. See RAID 0, RAID 1, RAID 2, RAID 4, RAID 5, and RAID 6.
RAID 4 - RAID Level 4: Dedicated Parity Drive: A commonly used implementation of RAID, provides block-level striping (like Level 0) with a parity disk. If a data disk fails, the parity data is used to create a replacement disk. A disadvantage to Level 4 is that the parity disk can create write bottlenecks. See RAID 0, RAID 1, RAID 2, RAID 3, RAID 5, and RAID 6.
RAID 5 - RAID Level 5: Block Interleaved Distributed Parity: Provides data striping at the byte level and also stripe error correction information. This results in excellent performance and good fault tolerance. Level 5 is one of the most popular implementations of RAID. See RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, and RAID 6.
RAID 6 - RAID Level 6: Independent Data Disks with Double Parity: Provides block-level striping with parity data distributed across all disks. See RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, and RAID 5.
RAID 0+1 - RAID Level 0+1: A Mirror of Stripes: Not one of the original RAID levels, two RAID 0 stripes are created, and a RAID 1 mirror is created over them. Used for both replicating and sharing data among disks.
RAID 10 - RAID Level 10: A Stripe of Mirrors: Not one of the original RAID levels, multiple RAID 1 mirrors are created, and a RAID 0 stripe is created over these.
RAID 7 - RAID Level 7: A trademark of Storage Computer Corporation (now effectively defunct) that adds caching to RAID levels 3 or 4. Patents were thrown out during suit against Hitachi (HDS).
RAID S - EMC Corporation's proprietary striped parity RAID system used in its Symmetrix storage systems.
RAIGE - RAID Across Independent Gigabit Ethernet: Term used by Pivot3. See RAID.
RAIN - Redundant Array of Independent Nodes: Term used by Adaptec. See RAID.
RAISE - Redundant Array of Independent Silicon Elements: Term invented by SandForce. which makes a SSD controller that uses this protection scheme. See RAID.
RAS - Reliability, Availability and Serviceability; computer hardware engineering term originally from IBM relating to running a long time without failure. This term is often used to loosely refer to bug reporting.
RBOD - RAID Data Protection JBOD Contrast with ABOD, EBOD, JBOD, SBOD and JBOK.
RCA - Root Cause Analysis; Reason behind a reported defect in a product
Reed-Solomon codes - RS - A Reed-Solomon encoder takes a block of digital data and adds extra "redundant" bits. Errors occuring during transmission or storage can be detected and corrected by using known polynomial coefficients.
In Reed-Solomon coding, source symbols are viewed as coefficients of a polynomial over a finite field. The original idea was to create n code symbols from k source symbols by oversamplying p(x) at n > k distinct points, transmit (or store) the sampled points, and use interpolation techniques at the receiver to recover the original message.
In practice today, RS codes are viewed as cyclic BCH codes, where encoding symbols are derived from the coefficients of a polynomial constructed by multiplying p(x) with a cyclic generator polynomial. This gives rise to efficient decoding algorithms.
RDMA - Remote Direct Memory Access; RDMA uses TCP/IP Etherenet and other network for high-performance server clustering eliminating the burden of excessive memory copies when communicating between servers. See RAID.
Repairspread - Lattus Safety Strategy (property of a storage policy) for objects. the Repairspread is the number of blockstores over which your data will be distributed upon. For instance, a spread width of 16 indicates that all data is spread over 16 different BlockStores. When an upload of a superblock fails, the repairspread attempts a new upload for that superblock, but with a different spread. The upload fails when no valid spread can be found.
repeater - A circuit that uses recovered clock to regnerate the outbound signal.
REPL - Run Eval Print Loop; Basic interactive debug and test environment for any interpretive command or script language.
REST - Representational State Transfer; Architectural style for communication on the World Wide Web where state is communicated in each message rather than being held by all communicants. A RESTful architecture consists of the elements: client-server, stateless server (no client context stored on server), cacheable unless noted otherwise, optionally layered invisibly, optional code on demand from servers for extensions to clients, uniform interface, identification of resources and related metadata, self-descriptive messages, and hypermedia.
retimer - A circuit that uses an independent clock to generate an outbound signal.
RFC - Request For Comments; Designation for most Internet standards, particularly protocol standards. RFCs are always associated with a unique number.
RIP - Router Information Protocol; Used on managed switches like Dell's Force10 S55.
RMON - Remote Monitoring; Used on managed switches like Dell's Force10 S55.
rNDC - Rack Network Daughter Card See NDC.
RoCE - RDMA over Converged Ethernet: Link layer protocol for fast communication.
RoHS - Restriction of Hazardous Substances; European Union compliance directive to restricts the use of hazardous materials found in electrical and electronic products.
Any RoHS complaint component is tested for the presence of Lead (Pb), Cadmium (Cd), Mercury (Hg), Hexavalent chromium (Hex-Cr), Polybrominated biphenyls (PBB), and Polybrominated diphenyl ether (PBDE). For cadmium and Hexavalent chromium, there must be less than 0.01% of the substance by weight at raw homogeneous materials level. For Leac, PBB, and PBDE, there must be no more than 0.1% of the material, when calculated by weight at raw homogeneous materials. Any RoHS complaitn component must have 100 ppm or less of mercury and the mercury must not have been intentially added to the component. In the EU, some military and medical equipment are excmpt for RoHS compliance.
rotational latency - The amount of delay in obtaining information from a disk drive that can be attributed to the rotation of the disk. For a disk rotating at 10,000 RPM, the average latency is 3 milliseconds.
RPM - Revolutions per Minute: Rotational speed of the media (e.g. disk), also known as the spindle speed.
RR - Round Robin (RR); Scheduling algorithm by which each ready device is served in sequence up to a maximum time splice. A device will not be served again until all other ready devices have had a chance to run. Contrast with MRU and Fixed.
RRDtool - Round-Robin Database Tool - Handles time-series data like network bandwidth, temperatures, CPU load, etc. Data are stored in a circular buffer based database. RRDtool can also refer to the graphical output, RRDgraph.
RSCN - Registered State Change Notification; a switch function that allows notification to registered nodes if a change occurs to other, specified nodes.
RSTP - Rapid Spanning Tree Protocol; Used on managed switches like Dell's Force10 S55.
RTS - Return to Stock
RV - Rotational Vibration: Hard disk drives often come with RV sensors to detect and avoid problems associated with rotational vibration such as poor seek times, read/write errors and even data loss.
S3 - Simple Storage Service; An online file storage web service. Amazon S3 provides storage through web services interfaces (REST, SOAP, and BitTorrent). Lattus offers a S3 interface API.
SA/A - Symmetric Active/Active Also see VMware.
SaaS - Software-as-a-Service Cloud Service Model; This is probably the most popular type of Cloud computing service model. Cloud Users just need to access, from the cloud clients, the software or application that the cloud providers installs in their cloud. SaaS usually involves virtual desktop, CRM (Customer Relationship Management), communications, emails, online games and the likes.
In this model, the software becomes the asset for businesses and consumers. This becomes one of the major advantage of a consumer. Because cloud users do not spend for infrastructure and the platforms. Consumers just need to use the software that the Cloud Providers provides. Thus reducing maintenance cost and risk of reducing ROI.
One perfect example of Saas is Gmail, a free email web application by Google similar to Yahoomail. The same email application can be accessed by multiple users in various locations. Since SaaS is elastic, it can clone tasks into different virtual machines whenever more and more users accesses Gmail. And then shrinks back when users signs out from Gmail.
Compare IaaS, PaaS, SaaS, HaaS, and GaaS.
SanDisk - Formerly SanDisk Corporation; Now a WD company. See http://www.sandisk.com.
SandForce - SandForce See http://www.sandforce.com.
SAS - Serial Attached SCSI; Storage access based on the SCSI command set.
A SAS drive utilizes the same form factor as a SATA drive but has several high performance advantages. First of all, there's the platter speed. While typical SATA drives operate at 7200 RPM, a SAS drive operates at 10K or 15K. Although the platter speed is double that of SATA, the MTBF remains at the industry standard of 1.2 million hours.
SAS drives are typically utilized in server and high-end workstation environments where speed and I/O frequency reign supreme.
SAS drives are linked directly to their controllers. SAS enables multiple devices (up to 128) of different sizes and types to be connected simultaneously with thinner and longer cables than traditional SCSI. Its full-duplex signal transmission supports 3.0Gb/s. In addition, SAS drives can be hot-plugged.
SAS devices can communicate with both SATA and SCSI devices (the backplanes of SAS devices are identical to SATA devices). A key difference between SCSI and SAS devices is the addition in SAS devices of two data ports, each of which resides in a different SAS domain. This enables complete failover redundancy as if one path fails, there is still communication along a separate and independent path.
Note: A SAS controller can run SAS and SATA drives. When a SAS controller runs SATA drives this is via SATA passthrough mode. A SATA controller can only run SATA drives (no SAS drives).
Mechanically, there is only one difference between SAS and SATA. While both use the same pinout for data and power connections, the two connectors are physically separated for SATA. For SAS, the two connector segments were merged, which makes it possible to attach a SAS drive to a SATA controller using the continuous connector, but you cannot hook up a SAS hard drive to a SATA controller using the SATA connector.
SAS supports multiple initiators, while SATA has no such provision.
SAS is dual ported whereas SATA is single ported. SAS is therefore capable of multi-path I/O without additional hardware. Moreover, SAS can make use of both ports to scale performance.SATA transfers data at half duplex while SAS can transfer data at full duplex. Thus, for SAS, each port can read and write data at the same time whereas SATA cannot.
SAS uses SCSI commands for error recovery and error reporting, which have more functionality than the ATA command set used by SATA.
SAS can support cables up to 8m long, while the maximum cable length for STA is 1m.
SAS1 - Serial Attached SCSI 1.0/1.1; SAS 1.0 and SAS 1.1 preserve legacy SCSI SATA compatibility, SAS1 has a line rate of 4 x 3 Gb/s (1.2 GB/s throughput). Maximum cable length is 8m. Storage features of SAS1 are RAID 6, small form factors, HPC, high capacity SAS drives, and ultra32D SCSI replacement.
See SAS2
SAS2 - Serial Attached SCSI 2.0; While SAS 2.0 does not preserve legacy SCSI SATA compatibility, it offers 3 Gb/s compatible with improved signaling, zoning management, and improved scalability. SAS 2.0 has a line rate of 4 x 6 Gb/s (2.4 GB/s throughput). Maximum cable length is 10m. Storage features of SAS2 RAS (data protection), security (FDE), clustering, larger topologies, SSDs, virtualization, external storage, and 4K sector size.
See SAS1
SAMBA - Open and free software re-implementation of the SMB/CIFS networking protocol from Microsoft Windows file sharing protocol.
SAN - Storage Area Netowrk; a network linking servers or workstations to disk arrays, tape-backup subsystems, and other devices, typically over fibre channel.
SanDisk Corp. - http://www.sandisk.com/
SATA - Serial ATA (SATA); Storage access based on the ATA command set.
SATA is a computer bus interface that connects host bus adapters to mass storage devices such as HDD and optical drives.
Note: A SAS controller can run SAS and SATA drives. When a SAS controller runs SATA drives this is via SATA passthrough mode. A SATA controller can only run SATA drives (no SAS drives).
SATM - Storage Array Type MEM; Prior to vSphere 4.1, this was called Storage Array Type Plugin (SATP). Also see VMware.
SATP - Storage Array Type Plugin; Starting with vSphere 4.1, the new name is Storage Array Teyp MEM (SATM). Also see VMware.
SBCCS - Single Byte Command Code Set Mapping
SBOD - Switched JBOD Contrast with ABOD, EBOD, JBOD, RBOD and JBOK.
scale out - The ability of something, especially a computer system, to adapt to increased demands. The term is often applied to storage to mean a system that is effectively unlimited in its ability to grow in size and access capabilities.
SCSI - Small Computer Systems Interface; both a protocol for transmitting large blocks of data and a parallel bus architecture.
SCSI-3 - A SCSI standard that defines transmission of SCSI protocol over serial links.
SDN Software Defined Network
Seagate Technology Public Limited Company - http://www.seagate.com/
sector - In disk technology, a sector is a 512 byte packet of data.
secure path - Secure Path is a load balancing, failover software package that HP optionally supplies with their storage arrays. Secure Path manages the paths via device manipuilation at a point before the OS creates disk devices for the user and thus applications see only one path.
SERDES - Serialing/deserializing circuitry; a circuit that converts a serial bit stream into parallel data characters, and parallel into serial.
serial - The transmission of data bits in sequential order over a single line.
server - A computer that processes end-user requests for data and/or applications.
SES - System Engineering Support See SPS.
SES - SCSI Enclosure Services; a subset of SCSI protocol used to monitor power, temperature, and fan status of enclosures.
The sg_ses (SCSI-Generic SES) tool is used to easily utilize these services. See sg.danny.dz/sg/index.html.
SFM - Oracle Service Fulfillment Manager; Program and service that keeps inventory information about hardware components.
See IB Trackable.
SFP - Small Form-factor Pluggable (optical transceiver module) (hardware description used with fiber connections)
SFTOS - Dell Force10 S-Series software. See FTOS.
SGI - Silicon Graphics, Incorporated
SIMM - Single Inline Memory Module; See DIMM.
single mode - A fiber-optic cabling specification that provides up to 10 kilometer distance between devices. Single mode fiber optic cable has a small diametral core that allows only one mode of light to propagate. Because of this, the number of light reflections created as the light passes through the core decreases, lowering attenuation and creating the ability for the signal to travel faster and further. Single mode is typically used in long distance telecommunications, cable TV, colleges and universities. Contrast with multimode
SKU - Stock Keeping Unit - A distinct type of item for sale (product or service) and all attributes associated with the item type that distinguish it from other item types. SKUs are not regulated nor standardized and are just the number used by a vendor for inventory management.
SLA - Service Level Agreement; Legal agreement to provide a predefined amount of capability.
SLC - Single Level Cell: Type of solid-state device where an individual cell of memory is used to store one bit of information. Constract SLC with SLTFS - Storage Library Tape File System
SMART - Self-Monitoring, Analysis and Reporting Technology; IBM originated design to monitor disk status by using various methods and devices (sensors). Today different manufacturers offer different measured values directly or indirectly related to hard disk health status and performance statistics. Most all modern ATA and SCSI disks offer S.M.A.R.T data.
SMAS - Server Management Command Line Protocol; Command line interface commonly used by iDRACs for their console support. Used by American Megatrends in the Lattus storage node ssh terminal connections.
SMB - Server Message Block protocol used by the Microsoft Windows network file system. See CIFS and SAMBA.
SMBIOS - System Management BIOS.
| Type | Description |
| 0 | BIOS Information |
| 1 | System Information |
| 2 | Baseboard (or Module) Information |
| 3 | System Enclosure or Chassis |
| 4 | Processor Information |
| 5 | Memory Controller Information (Obsolete) |
| 6 | Memory Module Information (Obsolete) |
| 7 | Cache Information |
| 8 | Port Connector Information |
| 9 | System Slots |
| 10 | On Board Devices Information |
| 11 | OEM Strings |
| 12 | System Configuration Options |
| 13 | BIOS Language Information |
| 14 | Group Associations |
| 15 | System Event Log |
| 16 | Physical Memory Array |
| 17 | Memory Device |
| 18 | 32-Bit Memory Error Information |
| 19 | Memory Array Mapped Address |
| 20 | Memory Device Mapped Address |
| 21 | Built-in Pointing Device |
| 22 | Portable Battery |
| 23 | System Reset |
| 24 | Hardware Security |
| 25 | System Power Controls |
| 26 | Voltage Probe |
| 27 | Cooling Device |
| 28 | Temperature Probe |
| 29 | Electrical Current Probe |
| 30 | Out-of-Band Remote Access |
| 31 | Boot Integrity Services (BIS) Entry Point |
| 32 | System Boot Information |
| 33 | 64-Bit Memory Error Information |
| 34 | Management Device |
| 35 | Management Device Component |
| 36 | Management Device Threshold Data |
| 37 | Memory Channel |
| 38 | IPMI Device Information |
| 39 | System Power Supply |
| 40 | Additional Information |
| 41 | Onboard Devices Extended Information |
| 42 | Management Controller Host Interface |
| 126 | Inactive |
| 127 | End-of-Table |
| 128-255 | Available for system- and OEM- specific information |
SMI - Structure of Management Information; a notation for setting or retrieving management variables over SNMP.
SMP - Symmetric Multi-Processor; Multiple CPUs of the exact same kind.
SMR - Shingled Magnetic Recording - Typically, disk drive tracks are separated by a gap. In SMR disk drives, the gap is removed to roughly double the density.
Write heads lay down a wider track than read heads need. In an SMR disk the write tracks are overlapped, leaving narrow tracks that are fine for the read head, but which can't be overwritten without destroying the data on adjacent tracks.
SMR is about 30% more capacity but slower and requires more host involvement.
In practice, the tracks on an SMR disk are laid down in bands of tracks. The band enables a partial rewrite of the disk by way of an expensive read-modify-write cycle like that used in RAID 5 arrays.
SMR drives tend to be sensitive to vibration.
See Kryder's Law.
SMTP - Simple Mail Transfer Protocol
SNIA - Storage Networking Industry Association: Standards association consisting of over 400 member storage companies. See http://www.snia.org
SNMP - Simple Network Management Protocol; a network management protocol designed to run over TCP/IP routed networks.
SNSM - StorNext Storage Manager
SNS - Simple Name Server; provided by a fabric switch, a service that simplifies discovery of devices.
SOAP - Simple Object Access Protocol; Protocol specification for exchaning structured information in the implementation of web services in computer networks. Uses XML for message format and relies on other application layer protocols (e.g. HTTP, SMTP) for message negotiation and transmission.
SOF - Start of Frame; a group of ordered sets that delineate the beginning of a frame. Contrast with EOF.
SOS - sosreport: Linux took to generate debugging information for the current system.
SPC-3 - The SPC-3 standard (2005) from INCITS
SpiderMonkey - Mozilla's JavaScript engine written in C/C++.
SPS - Sales Professional Services; Field support team for customers who purchased support. See SES.
SR - Service Request; An SR or customer request for problem/inquiry resolution via the service organization that relates in some way to the product they have purchased.
SRAM - Static Random Access Memory; memory that stores data using static charges that maintain their state only as long as electric power is supplied. SRAM differs from DRAM which stores data dynamically and constantly needs to refresh the data stored in memory. Because SRAM stores data statically, it is faster and requires less power than DRAM. However, SRAM is more expensive to manufacture than DRAM because of its more complex structure. Because of the complexity, SRAM cannot hold as much data as DRAM. SRAM is commonly used in smaller applications such as cache memory and hard drive buffers. SRAM is a type of NVRAM.
SRM - Storage Resource Management; management of disk volumes and file resources.
SSD - Solid State Device
ssh - Secure Shell; Telecommunication using Unix shell commands over an encrypted communication link.
SSL - Secure Socket Layer See TLS.
StableTractm - WD name for a disk drive in which the motor shaft is secured at both ends to reduce system-induced vibration and stabilize platters for accurate tracking, during read and write operations.
staggered spinup - SATA feature that allows the system to control whether the disk drive will spin up immediately or wait until the interface is fully ready.
STEC - sTec, Inc. is a computer data storage company headquartered in California. The company designs, develops and manufactures solid-state drives based on flash memory and dynamic random access memory (DRAM) for large-scale data center environments and OEM customers. Formerly the company was Simple Technology.
storage - Any device used to store data; typically, magnetic disk media or tape.
store-and-forward - A switching technique that requires buffering an entire frame before a routing decision is made.
STP - Spanning Tree Protocol; Used on managed switches like Dell's Force10 S55.
StorNext - Quantum Corporation's metadata controller for scale-out storage management.
striping - Typically a RAID technique for writing a single file to multiple disks on a block-by-block basis.
The main performance-limiting issues with disk storage relate to the slow mechanical components that are used for positioning and transferring data. Since a RAID array has many drives in it, an opportunity presents itself to improve performance by using the hardware in all these drives in parallel. For example, if one needs to read a large file, instead of pulling it all from a single hard disk, it is much faster to chop it up into pieces, store some of the pieces on each of the drives in an array, and then use all the disks to read back the file when needed. This technique is called striping, after the pattern that might be visible if you could see these "chopped up pieces" on the various drives with a different color used for each file. It is similar in concept to the memory performance-enhancing technique called interleaving.
Striping can be done at the byte level, or in blocks. Byte-level striping means that the file is broken into "byte-sized pieces." The first byte of the file is sent to the first drive, then the second to the second drive, and so on. Sometimes byte-level striping is done as a sector of 512 bytes. Block-level striping means that each file is split into blocks of a certain size and those are distributed to the various drives. The size of the blocks used is also called the stripe size (or block size, or several other names), and can be selected from a variety of choices when the array is set up.
The stripe width is the number of parallel stripes that can be written to or read from simultaneously. This is equal to the number of disks in the array.
The stripe size of the array is the size of the stripes written to each disk. This is sometimes referred to as block size, chunk size, stripe length or granularity.
RAID arrays that stripe in blocks typically allow the selection of block sizes in kilobytes ranging from 2 kB to 512 kB or more in powers of two. Byte-level striping (as in RAID 3) uses a stripe size of one byte or perhaps a small number like 512, usually not selectable by the user.
subscription-manager - Red Hat subscription-base update tool. Older versions of Red Hat used a tool named up2date. See yum and apt-get.
Superblock - For Lattus, a superblock is the logical storage unit for the BitSpread backend. This unit is not visible to the end user. When storing an object in BitSpread, the client divides the object into one or multiple superlocks before encoding the data itself.
Superblocks are to be consider the logical storage unit of the storage backend. They are not visible to the API, which makes requests based on an object-namespace pair.
the maximum size is 64 MB but that is configurable from 1 MB to 64 MB. Superblocks are assigned on a per policy basis and all superblocks of a single object will have the same blockstore spread.
Cache Daemons cache superblocks for requently read objects.
Repairspread is a safety strategy property of a storage policy. When an upload of a superblock fails, the Reparispared attempts a new upload for that upserblock, but with a different spread. The upload fails when no valid spread can be found.
In contrast, for a file system, a superblock is a record of the characteristics of a filesystem including its size, the block size, the empty and the filled blocks and their respective counts, the size and location of the inode tables, the disk block map and usage information and the size of the block groups.
Swift - OpenStack Object Storage is code named Swift. It is a scalable redundant object storage system.
switch - In fibre channel, a device providing full bandwidth per port and high-speed routing of data via link-level addressing.
Symform - Quantum's newly acquired (Aug. 2014) cloud storage platform.
T10 - INCITS Technical Committee T10 - SCSI Storage Interfaces. See http://www.t10.org.
T11 - INCITS Technical Committee T11 - Fibre Channel Interfaces. See http://www.t11.org. Also see Physical Variants: T11.2 and Fibre Channel Interconnection Schemes: T11.3.
T11.2 - INCITS Technical Committee T11.2 - Fibre Channel Interfaces: Physical Variants. See http://www.incits.org/committees/t11.2.
T11.3 - INCITS Technical Committee T11.3 - Fibre Channel Interfaces: Fibre Channel Interconnection Schemes. See http://www.incits.org/committees/t11.3.
T13 - INCITS Technical Committee T13 - ATA Storage Interfaces. See http://www.t13.org.
TAM - Total Addressable Market (also Total Available Market or Total Annual Market): used to describe the revenue opportunity available for a product or service.
target - Typically a disk array or a tape subsystem on a fibre channel network.
TCO - Total Cost of Ownership (includes such things as purchase price, power, cooling, licensing, upgrading, personnel training, supporting maintenance, resale value)
TDMR - Two Dimensional Magnetic Recording
TCP/IP - Transmission Control Protocol over Internet Protocol.
telnet - A virtual terminal emulation facility over TCP/IP.
tenancy - Possession of an FC-AL by a device to conduct a transaction.
Teradata Corporation - http://www.teradata.com/?LangType=1033
TLC - Triple Level Cell: Three bits per cell for eight total states in solid state memory. See MLC
TLER - Time Limited Error Recovery: If there is a disk error, some systems may take 20 seconds to recover so you need TLER. Most NAS assume that after 7 seconds, the entire drive has failed and starts rebuilding.
TLS - Transport Layer Security See SSL.
TOI - Transfer Of Information; training session
topology - The physical or logical arrangement of devices in a networked configuration.
TPC - Third Party Copy; a protocol for performing tape backups without consuming server resources.
TPGS - Target Port Group Support; a method to enable a new storage device to be automatically detected and to allow methodologies to understand different port characteristics and failover behaviors; also provides a method for determining the access characteristics of a path to a logical unti through a target port. See ALUA.
TPM - Trusted Platform Module; A chip on the motherboard used to generate keys based on system files for disk encryption.
transceiver - A device that converts one form of signaling to another for both transmission and reception; in fiber optics, conversion from optical to electrical.
TRAMP - Transparent Remote (file) Access, Multiple Protocol; Extension of directory path syntax to include FTP, SSH, Rsync, Sudo and other protocols for accessing and manipulating remote files and connections. TRAMP is commonly used with EMACS. Some examples are /ssh:user@host:mydirectory/myfile.ext, /ftp:ftpuser@ftp.example.com, /sudo::/etc/hosts.
trap - A SNMP trap is an unsolicited message sent to a notification receiver. Unlike messages sent to a SNMP agent, traps are not confirmed by the receiver. If it is critical that a Network Management System (NMS) be able to detect when a message is lost, then the device must provide a method for detecting lost messages, such as a log table or sequence number. If the device supports informs, critical notifications should be sent using informs.
TSB - Technical Support Bulletin
TTLS - Tunneled Transport Layer Security
TUI - Text User Interface
UFM - Unified Fabric Manager - Mellanox cluster manager designed for high availability and reliability.
VAR - Value-Added Reseller; a company that sells combined and/or enhanced products manufactured by other companies.
varbind - Variable Binding: A SNMP varbind is an OID for an object and the data value associated with that object.
variable capacity - A variable capacity disk drive is one selected out of a bin of drives that average the advertised capacity. Each particular drive may have a slightly higher or lower capacity. By allowing for variable capacity, more usable platters/drives can be shipped from the factory overall (lowering the reject rate and thereby lowering the average price).
The variable size is well suited for archival storage systems that do not use conventional RAID (where an entire array of matching capacity disks are spinning simultaneously). Drives are spun up and written to individually, or spun up individually to service the occasional read request. This saves power overall, and it also means the individual drivesw can vary in capacity with no ill effects.
Contrast this with the more traditional fixed capacity disk drive. An example of a variable capacity drive would be the WD AE HDD series of drives.
Veritas - Veritas Software is a software company acquired by Symantec. Veritas is known for its storage management software including the first journaling file system, VxFS, VxVM, VCS, and NetBackup.
VCSEL - Vertical Cavity Surface Emitting Laser: Unlike edge-emitting CD lasers, vertical cavity lasers emit light perpendicular to the surface of the wafer substrate. Compared to CD lasers, they consume less power, radiate less heat, and maintain calibration better.
Violin Memory Inc. - http://www.teradata.com/?LangType=1033
VirtualBox - Oracle VM VirtualBox is virtualization software package. See http://www.virtualbox.org.
VPI - Virtual Protocol Interconnect - Mellanox switch protocol to allow seamless InfiniBand and Ethernet connectivity.
VLAN - Virtual Local Area Network; A non-physical (software or switch firmware) supported group of networking devices in the same broadcast domein. Although layer 2 switches can only break up collision domains but VLANs can be used to break up broadcast domains. When using VLANs in network that have multiple interconnected switches, you need to use VLAN trunking between the switches. See http://www.9tut.com/virtual-local-area-network-vlan-tutorial
VM - Virtualization Management / Virtual Manager
VMkernel - Part of the ESX(i) product that manages interactions with the devices, handles memory allocation, and schedules access to the CPU resources, and other things such as PSA.
VMware - VMware, Inc. is a software company acquired by EMC that makes cloud and virtualzation software. Its enterprise software hypervisors are VMware ESX and VMware ESXi. See http://www.vmware.com.
VPC - Virtual Private Cloud, a private cloud existing within a shared or public cloud
VPN - Virtual Private Network, a private network that interconnects remote networks through primarily public communication infrastructures such as the Internet.
VPS - Virtual Private Server, a term used by Internet hosting services to refer to a virtual machine.
VRRP - Virtual Router Redundancy Protocol; Used on managed switches like Dell's Force10 S55.
vSphere - VMware's cloud computing virtualization operating system.
WAN - Wide Area Network; a network linking geographically remote sites.
Western Digital Corporation - WD - http://www.wdc.com
World-Wide Name - A registered, unique 64-bit identifier assigned to nodes and ports.
WOS - Web Object Scaler; DDN WOS is an object-based storage system that doesn't use a file system.
See DDN.
WWD - Universal Worldwide Device Identifier (WWD ID); a method of uniquely naming devices. See Veritas.
Xcellis - StorNext file and data management hardware and software product from Quantum. Xcellis provides multi-tier data storage with rapid indexed storage and retrieval for multiple types of workstations via Fibre channel, NAS and DLC protocols.
xHCI - Extensible Host Controller Interface; xHCI Pre-Boot Driver. This controls the handling of USB3. xHCI is the USB 3.0 controller and EHCI is the USB 2.0 controller. Usually a BIOS setting that takes one of Smart Auto, Auto, Enabled, or Disabled. For the xHCI Pre-Boot Driver, when Enabled: The USB 3.0 ports are routed to the xHCI controller before booting to OS (default). When Disabled: The USB 3.0 ports are routed to the EHCI controller before booting to OS. When set to Enabled, the xHCI mode will be automatically set to Smart Auto. When set to Disabled, the xHCI will be automatcially set to Auto.
yum - Yellowdog Updater, Modified - A software package manager and installer similar to up2date (Red Hat) or apt (Debian) that keeps a database of knowledge of software packages and where on the Internet to get them. Note that dnf (stands for nothing officially) is replacing yum in Fedora 22 and soon in other Red Hat releases. See subscription-manager and apt-get.
zoning - Provided by fabric switches or hubs, a function that allows segregation of node by physical port, name, or address.
3M - Minnesota Mining and Manufacturing - http://www.3m.com
802.2 - IEEE 802.2 Logical Link Control (LLC) - a.k.a. ISO/IEC 8802-2 - The LLC sublayer presents a uniform interface to the user of the data link service. The LLC header contains the logical address of the network layer entity that created the message, the logical addresses of the network layer entity intended to receive the message and control information (e.g. flow control). See http://www.ieee802.org/2/ and https://en.wikipedia.org/wiki/IEEE_802.2
Kastner Home Page
Last modified February 22, 2023
webmaster@kastner.us