0
InfiniBand® is a network communications and storage system that makes use of high-speed serial links and processors to address the concerns of performance and reliability for high-performance computing. It is capable of addressing 64,000 nodes, and may provide speeds up to 2.5 gigabits per second (Gbps). This speed is about 2,500 million bits per second (Mbps), which may download a 1-gigabitfile in less than 4 seconds.
What else can I help you with?
What are the disadvantages of infiniband?
1)complex in design 2)few platform supports
What are the name of supercomputer?
it all depends on what super computer your onabout really .... here's a few super computers : PARAM ( India ) IBM ( america ) fujitsu ( im not sure ) EKA ( india ) Infiniband ( america ) graph 500 ( Europe ) IBM Blue Gene/Q ( america ) Sequioa ( england )
What is the ram capacity of roadrunner IBM supercomputer?
Overall system information: 6,480 Opteron processors with 51.8 TiB RAM (in 3,240 LS21 blades)216 System x3755 I/O nodes26 288-port ISR2012 Infiniband 4x DDR switches296 racks2.35 MW power search it on wikipedia for the rest.
How many super computers in India?
* Four new entries in the list! * Total of 13 entries in the list. * The performance criteria is minimum of 1.259 TFlops. * First 4 in the list are also in Top 500. * The average performance is 18.75 TFlops. The combined performance is 243.71 TFlops. * 11 of the systems are from HP. Others are from IBM. * City-wise distribution: Bangalore - 4, Pune - 3, Chennai - 3, Delhi - 1, Mumbai -1, Hyderabad - 1. * 5 of the systems are from Centre for Development of Advance Computing (CDAC), proving its status as a leading high performance computing center in the nation. Rank Site System Cores/Processor Sockets/Nodes Rmax (TFlops) Rpeak (TFlops) 1 Computational Research Laboratories Ltd, Pune HP Cluster Platform 3000 BL460c (Dual Intel Xeon 3 GHz quad core E5365 (Clovertown) w/Infiniband 4X DDR) 14400/3600/1800 132.80 172.60 2 Center for Development of Advanced Computing (C-DAC) India PARAMcluster (Intel Xeon (Tigerton) 2.93 GHz quad core quad processor X73xx nodes w/Infiniband) 4608/1152/288 37.80 54.01 3 Indian Institute of Science, Bangalore IBM eServer Blue Gene Solution, BlueGene/L (IBM PowerPC 700 MHz 440x5 processors w/Proprietary Interconnect) 8196/4096/4096 17.81 22.94 4 Paprikaas Interactive Services India BL460c (Xeon 54xx 3.0GHz w/GigEthernet) 1920/ / 13.16 23.04 5 Tata Institute of Fundamental Research (TIFR), Mumbai IBM Blue Gene Solution, BlueGene/P (IBM PowerPC 850 MHz 450 processors w/Proprietary Interconnect) 4092/1024/1024 11.32 13.5 6 Indian Institute of Technology Madras, Chennai HP 256 Intel Xeon Processor E5472 (3 GHz w/Infiniband) 1024/256/128 8.62 12.29 7 Indian Institute of Technology Madras, Chennai HP 256 Intel Xeon Processor E5472 (3 GHz w/Gigabit) 1024/256/128 4.55 8.64 8 Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore HP Proliant DL140 G3 (dual processor dual core Intel Xeon 3GHz 5160 nodes w/Infiniband 4X DDR) 512/256/128 3.86 6.00 9 Institute of Genomics and Integrative Biology, Delhi HP CP3000 (576 Intel Xeon Processor X3.6GHz/800-2MB w/Infiniband) 576/576/288 3.06 4.08 10 Centre for Development of Advanced Computing (CDAC), Chennai HP Cluster (Dual Intel Quad Core Xeon EM64 processor w/Infiniband) 320/80/40 2.999 3.943 11 Centre for Development of Advanced Computing (CDAC), Bangalore HP Cluster (Dual Intel Quad Core Xeon EM64 processor w/Infiniband) 320/80/40 2.976 4.044 12 Centre for Development of Advanced Computing (CDAC), Hyderabad HP Cluster (Dual Intel Quad Core Xeon EM64 processor w/Infiniband) 320/80/40 2.614 3.94 13 Centre for Development of Advanced Computing (CDAC), Pune HP-DL580-G5 (Intel Xeon (Tigerton) 2.933 GHz quad core quad processor X7350 nodes w/10Gbps PARAMNet-3) 256/64/16 2.165 3.00
Which type of hard drive uses 8b10b encoding scheme to encode and decode data sent along the cable?
Currently these are the only areas that use 8B/10B encoding: PCI Express * IEEE 1394b * Serial ATA * SAS * Fibre Channel * SSA * Gigabit Ethernet (except for the twisted pair based 1000Base-T) * InfiniBand * XAUI * Serial RapidIO * DVI and HDMI (Transition Minimized Differential Signaling) * DVB Asynchronous Serial Interface (ASI) * DisplayPort Main Link * HyperTransport * Common Public Radio Interface (CPRI) * USB 3.0
What exteranl is faster than USB or Firewire?
There are multiple devices faster than USB and Firewire, but for the consumer who uses regular hardware, I would recommend the following three interfaces in the order of the cost they carry. 1. USB3 2. eSATA 3. a Storage Managed/Backup device which makes a massive hard drive (or RAID Array) availble over Ethernet on CIFS/NFS. To Answer your question, if you were a corporation with some deeper pocketbooks, you could use: 1. Iscsi - Scsi over Ethernet 2. FibreChanel - SAN Storage over Fiber. 3. Infiniband - proprietary interface, but very quick. 4. A whole host of more expensive proprietary and open source standards to access multiple raid arrays (sometimes in parallel), for specialized applications [Google HPC Cluster Filesystems] and Filesystems interconnects
What is a corded?
cheese whiz and water bottles Fiber Optic Patch Cord is used to interconnect Transmission equipment to the Fiber optic patch panel. SAILING is a professional manufacturer of Fiber Optic Patch Cord in China. The Patch Cord made by SAILING supports longer distance with lower loss for gigabit Ethernet, fiber channel, infiniband ATM, and internet protocol applications, and it is Specifically designed for use with today's narrower aperture components. The patented injection molding process provides each connection greater durability in resisting pulls, strains and impacts from cabling installs. On the other hand, each cable is 100% optically inspected and tested for insertion loss before you receive it, and a pull-proof jacket design surrounds the popular 50/125 multimode fiber, immune to electrical interference Applications: Telecom and Datacom Storage Network CATV $ Multimedia Application Systems Integration for Long Haul, Metro and Access Network Fiber to the Indoor (FTTX) Test Local Area Network(LAN)
What is optic cord?
Fiber Optic Patch Cord is used to interconnect Transmission equipment to the Fiber optic patch panel. SAILING is a professional manufacturer of Fiber Optic Patch Cord in China. The Patch Cord made by SAILING supports longer distance with lower loss for gigabit Ethernet, fiber channel, infiniband ATM, and internet protocol applications, and it is Specifically designed for use with today's narrower aperture components. The patented injection molding process provides each connection greater durability in resisting pulls, strains and impacts from cabling installs. On the other hand, each cable is 100% optically inspected and tested for insertion loss before you receive it, and a pull-proof jacket design surrounds the popular 50/125 multimode fiber, immune to electrical interference Applications: Telecom and Datacom Storage Network CATV $ Multimedia Application Systems Integration for Long Haul, Metro and Access Network Fiber to the Indoor (FTTX) Test Local Area Network(LAN)
What kinds of storage devices do large computer systems use?
Large computer systems usually use purpose specific storage based on the requirements of the system. For instance, a large file server would use several racks of simple storage servers usually via something like iSCSI over fiber or Infiniband. A large super computer may not need that much storage, but it would need to be faster, so there would be large RAM requirements with little permanent storage to speak of. A Cray CX1 Supercomputer can have up to 64 Processors interconnected, but while a large computing system, usually has less need for disk space. The HP NonStop OS for highly available "never go offline" kind of systems are used for things like certain stock exchanges, and have both a large number of storage disks and a large amount of RAM in most applications. Specific Disk types are traditionally SCSI or SAS drives, but the rise of SATA speeds and reliablity, coupled with the much cheaper cost means many storage systems are moving to the new SATA standards. SCSI has higher reliability, but if an SATA drive can store more, and for a fourth of the cost, it's sometimes more cost effective to create a storage array that can stand to lose several drives from a RAID array, and simply have disks on hand to replace them as they go. SCSI is still used in situations where such maintenance is not much of an option.
Definition of primary memory?
Primary storage, also known as main storage or memory, is the main area in a computer in which data is stored for quick access by the computer's processor. On today's smaller computers, especially personal computers and workstations, the term random access memory (RAM) - or just memory - is used instead of primary or main storage, and the hard disk, diskette, CD, and DVD collectively describe secondary storage or auxiliary storage.The terms main storage and auxiliary storageoriginated in the days of the mainframecomputer to distinguish the more immediately accessible data storage from storage that required input/output operations. An earlier term for main storage was core in the days when the main data storage contained ferrite cores.2) Primary storage is sometimes used to mean storage for data that is in active use in contrast to storage that is used for backup purposes. In this usage, primary storage is mainly the secondary storage referred to in meaning 1. (It should be noted that, although these two meanings conflict, the appropriate meaning is usually apparent from the context.)RELATED GLOSSARY TERMS: hard disk, magnetoresistive head technology, yottabyte, serverless backup, byte, partition, InfiniBand, failover, RAMAC (random access method of accounting and control), Fibre Channel
What is the difference between an Ethernet Address and an mac address?
Ethernet address is really not a good term to use. I think you mean a MAC address. A MAC address is given to any device that accesses the internet. Such devices include WiFi cards, modems, and routers. These are only used by networks that devices connect to, and not by websites. They can be used for filtering and identifying devices on a network.Ip (Internet Protocol) addresses are assigned to networks (one per network) by an ISP (Internet Service Provider) such as Comcast. These are sent to everything you interact with online, so that your network, rather than your computer, is identified.Now there are two kinds of Ip addresses: internal and external. Internal are used by your network to identify each device that connects to it. External Ip's are assigned by your ISP to identify you to the world. I explained the difference more in-depth in another answer, which is right here:What_is_the_difference_between_an_Ip_address_and_a_http_address
