InfiniBand

From HandWiki
Short description: High-speed, low-latency computer networking bus used in supercomputing

InfiniBand (IB) is a computer networking communications standard used in high-performance computing that features very high throughput and very low latency. It is used for data interconnect both among and within computers. InfiniBand is also used as either a direct or switched interconnect between servers and storage systems, as well as an interconnect between storage systems. It is designed to be scalable and uses a switched fabric network topology.

(As of 2014), it was the most commonly used interconnect in supercomputers. In 2016, Ethernet replaced InfiniBand as the most popular system interconnect of TOP500 supercomputers.[1]

Mellanox (now Nvidia Networking) manufactures InfiniBand host bus adapters and network switches, and, in February 2016, it was reported that Oracle Corporation had engineered its own InfiniBand switch units and server adapter chips for use in its own product lines and by third parties.[2] Mellanox IB cards are available for Solaris, FreeBSD,[3][4] RHEL, SLES, Windows, HP-UX, VMware ESX,[5] and AIX.[6]

As an interconnect, IB competes with Ethernet, Fibre Channel, and Intel Omni-Path.

The technology is promoted by the InfiniBand Trade Association.

Specification

Performance

Characteristics
  SDR DDR QDR FDR10 FDR EDR HDR NDR XDR
Signaling rate (Gbit/s) 0002.5 0005 0010 0010.3125 0014.0625[7] 0025.78125 0050 0100 0250
Theoretical
effective
throughput
(Gb/s)[8] 		for 01 links 0002 0004 0008 0010 0013.64 0025 0050 0100 0250
for 04 links 0008 0016 0032 0040 0054.54 0100 0200 0400 1000
for 08 links 0016 0032 0064 0080 0109.08 0200 0400 0800 2000
for 12 links 0024 0048 0096 0120 0163.64 0300 0600 1200 3000
Encoding (bits) 8b/10b 64b/66b PAM4 t.b.d.
Adapter latency (µs)[9] 0005 0002.5 0001.3 0000.7 0000.7 0000.5 less? t.b.d. t.b.d.
Year[10] 2001, 2003 2005 2007 2011 2011 2014[11] 2018[11] 2021[11] after 2023?

Links can be aggregated: most systems use a 4× aggregate. 8× and 12× links are typically used for cluster and supercomputer interconnects and for inter-switch connections.

InfiniBand also provides RDMA capabilities for low CPU overhead.

Topology

InfiniBand uses a switched fabric topology, as opposed to early shared medium Ethernet. All transmissions begin or end at a channel adapter. Each processor contains a host channel adapter (HCA) and each peripheral has a target channel adapter (TCA). These adapters can also exchange information for security or quality of service (QoS).

Messages

InfiniBand transmits data in packets of up to 4 KB that are taken together to form a message. A message can be:

Physical interconnection

InfiniBand switch with CX4/SFF-8470 connectors

In addition to a board form factor connection, it can use both active and passive copper (up to 10 meters) and optical fiber cable (up to 10 km).[12] QSFP connectors are used.

The InfiniBand Association also specified the CXP connector system for speeds up to 120 Gbit/s over copper, active optical cables, and optical transceivers using parallel multi-mode fiber cables with 24-fiber MPO connectors.[citation needed]

API

InfiniBand has no standard API. The standard only lists a set of verbs such as ibv_open_device or ibv_post_send, which are abstract representations of functions or methods that must exist. The syntax of these functions is left to the vendors. Sometimes for reference this is called the verbs API. The de facto standard software stack is developed by OpenFabrics Alliance. It is released under two licenses GPL2 or BSD license for Linux and FreeBSD, and as Mellanox OFED for Windows (product names: WinOF / WinOF-2; attributed as host controller driver for matching specific ConnectX 3 to 5 devices)[13] under a choice of BSD license for Windows. It has been adopted by most of the InfiniBand vendors, for Linux, FreeBSD, and Windows. IBM states this at their knowledge center on verbs API:

An AIX® application can determine the verbs API that is either the Open Fabrics Enterprise Distribution (OFED) verbs or the AIX InfiniBand (IB) verbs that must communicate with a specific destination.[14]

A presentation from Mellanox Technologies, dated 2014, with title "Verbs programming tutorial" states on page 31:[15]

  • libibverbs, developed and maintained by Roland Dreier since 2006, are de facto the verbs API standard in *nix
    • Developed as an Open source
    • The kernel part of the verbs is integrated in the Linux kernel since 2005 – Kernel 2.6.11
    • Inbox in several *nix distributions
    • There are level low-level libraries from several HW vendors

History

InfiniBand originated in 1999 from the merger of two competing designs: Future I/O and Next Generation I/O. This led to the formation of the InfiniBand Trade Association (IBTA), which included Compaq, Dell, Hewlett-Packard, IBM, Intel, Microsoft, and Sun. At the time it was thought some of the more powerful computers were approaching the interconnect bottleneck of the PCI bus, in spite of upgrades like PCI-X.[16] Version 1.0 of the InfiniBand Architecture Specification was released in 2000. Initially the IBTA vision for IB was simultaneously a replacement for PCI in I/O, Ethernet in the machine room, cluster interconnect and Fibre Channel. IBTA also envisaged decomposing server hardware on an IB fabric. Following the burst of the dot-com bubble there was hesitation in the industry to invest in such a far-reaching technology jump.[17]

Timeline

  • 2001: Mellanox ships InfiniBridge 10Gbit/s devices and ships over 10,000 InfiniBand ports.[18]
  • 2002: adoption takes a setback when Intel announces that instead of shipping IB chips it would focus on developing PCI Express, and Microsoft discontinues IB development in favor of extending Ethernet, although Sun and Hitachi continue to support IB.[19]
  • 2003: Virginia Tech builds an InfiniBand cluster ranked number three on the Top500 at the time.
  • 2004: IB starts to see adoption as a clustering interconnect, beating Ethernet on latency and price.[17] The OpenFabrics Alliance develops a standardized, Linux-based InfiniBand software stack. The following year Linux adds IB support.[20]
  • 2005: IB begins to be implemented as an interconnect for storage devices.[21]
  • 2009: of the top 500 supercomputers in the world, Gigabit Ethernet is the internal interconnect technology in 259 installations, compared with 181 using InfiniBand.[22]
  • 2010: Market leaders Mellanox and Voltaire merge, leaving just one other IB vendor, QLogic, primarily a Fibre Channel vendor.[23] Oracle makes a major investment in Mellanox.
  • 2011: FDR switches and adapters announced at the International Supercomputing Conference.[24]
  • 2012: Intel acquires QLogic's InfiniBand technology.[25]
  • 2016: Oracle Corporation manufactures its own InfiniBand interconnect chips and switch units.[26]
  • 2019: Nvidia acquired Mellanox for $6.9B[27]

Ethernet over InfiniBand

Ethernet over InfiniBand, abbreviated to EoIB, is an Ethernet implementation over the InfiniBand protocol and connector technology.

EoIB enables multiple Ethernet bandwidths varying on the InfiniBand (IB) version.

Ethernet's implementation of The Internet Protocol Suite, usually referred to as TCP/IP, is different to some of the implementations used on top of the InfiniBand protocol in IP over IB (IPoIB).

Ethernet over InfiniBand performance
Type Lanes Bandwidth (Gbit/s) Compatible Ethernet type(s) Compatible Ethernet quantity
SDR 001 0002.5 GbE to 2.5 GbE 02 × GbE to 1 × 02.5 GbE
004 0010 GbE to 10 GbE 10 × GbE to 1 × 10 GbE
008 0020 GbE to 10 GbE 20 × GbE to 2 × 10 GbE
012 0030 GbE to 25 GbE 30 × GbE to 1 × 25 GbE + 1 × 05 GbE
DDR 001 0005 GbE to 5 GbE 05 × GbE to 1 × 05 GbE
004 0020 GbE to 10 GbE 20 × GbE to 2 × 10 GbE
008 0040 GbE to 40 GbE 40 × GbE to 1 × 40 GbE
012 0060 GbE to 50 GbE 60 × GbE to 1 × 50 GbE + 1 × 10 GbE
QDR 001 0010 GbE to 10 GbE 10 × GbE to 1 × 10 GbE
004 0040 GbE to 40 GbE 40 × GbE to 1 × 40 GbE

See also

References

  1. ↑ "HIGHLIGHTS – JUNE 2016". June 2016. https://www.top500.org/lists/top500/2016/06/highlights/. "InfiniBand technology is now found on 205 systems, down from 235 systems, and is now the second most-used internal system interconnect technology. Gigabit Ethernet has risen to 218 systems up from 182 systems, in large part thanks to 176 systems now using 10G interfaces." 
  2. ↑ "Oracle Engineers Its Own InfiniBand Interconnects". 23 February 2016. http://www.nextplatform.com/2016/02/22/oracle-engineers-its-own-infiniband-interconnects/. 
  3. ↑ "Mellanox OFED for FreeBSD". Mellanox. http://www.mellanox.com/page/products_dyn?product_family=193. 
  4. ↑ "FreeBSD Kernel Interfaces Manual, mlx5en" (in en). FreeBSD. 3 December 2015. https://www.freebsd.org/cgi/man.cgi?query=mlx5en. 
  5. ↑ "InfiniBand Cards - Overview". Mellanox. http://www.mellanox.com/page/infiniband_cards_overview. 
  6. ↑ "Implementing InfiniBand on IBM System p (IBM Redbook SG24-7351-00)". http://www.redbooks.ibm.com/redbooks/pdfs/sg247351.pdf. 
  7. ↑ https://cw.infinibandta.org/document/dl/7260
  8. ↑ "Archived copy". http://www.infinibandta.org/content/pages.php?pg=technology_overview. 
  9. ↑ http://www.hpcadvisorycouncil.com/events/2014/swiss-workshop/presos/Day_1/1_Mellanox.pdf // Mellanox
  10. ↑ Panda, Dhabaleswar K.; Sayantan Sur (2011). "Network Speed Acceleration with IB and HSE". Designing Cloud and Grid Computing Systems with InfiniBand and High-Speed Ethernet. Newport Beach, CA, USA: CCGrid 2011. pp. 23. http://www.ics.uci.edu/~ccgrid11/files/ccgrid11-ib-hse_last.pdf#page=23. 
  11. ↑ 11.0 11.1 11.2 "InfiniBand Roadmap - Advancing InfiniBand" (in en-US). https://www.infinibandta.org/infiniband-roadmap/. 
  12. ↑ "Specification FAQ". ITA. http://www.infinibandta.org/content/pages.php?pg=technology_faq. 
  13. ↑ Mellanox OFED for Windows - WinOF / WinOF-2
  14. ↑ Verbs API
  15. ↑ Verbs programming tutorial
  16. ↑ Pentakalos, Odysseas. "An Introduction to the InfiniBand Architecture". http://www.oreillynet.com/pub/a/network/2002/02/04/windows.html. 
  17. ↑ 17.0 17.1 Kim, Ted. "Brief History of InfiniBand: Hype to Pragmatism". Oracle. https://blogs.oracle.com/RandomDude/entry/history_hype_to_pragmatism. 
  18. ↑ "Timeline". Mellanox Technologies. http://www.mellanox.com/page/timeline. 
  19. ↑ "Sun confirms commitment to InfiniBand". https://www.theregister.co.uk/2002/12/30/sun_confirms_commitment_to_infiniband/. 
  20. ↑ "Linux Kernel 2.6.11 Supports InfiniBand". http://www.internetnews.com/dev-news/article.php/3485401. 
  21. ↑ "Is InfiniBand poised for a comeback?", Infostor 10 (2), http://www.infostor.com/index/articles/display/248655/articles/infostor/volume-10/issue-2/news-analysis-trends/news-analysis-trends/is-infiniband-poised-for-a-comeback.html 
  22. ↑ Lawson, Stephen. "Two rival supercomputers duke it out for top spot". COMPUTERWORLD. http://news.idg.no/cw/art.cfm?id=FB70C2C5-1A64-6A71-CEEA6C17D51B1E3C. 
  23. ↑ Raffo, Dave. "Largest InfiniBand vendors merge; eye converged networks". http://itknowledgeexchange.techtarget.com/storage-soup/largest-infiniband-vendors-merge-eye-converged-networks/. 
  24. ↑ "Mellanox Demos Souped-Up Version of InfiniBand". CIO. 20 June 2011. http://www.cio.com/article/684732/Mellanox_Demos_Souped_Up_Version_of_Infiniband. 
  25. ↑ "Intel Snaps Up InfiniBand Technology, Product Line from QLogic". HPCwire. January 23, 2012. https://www.hpcwire.com/2012/01/23/intel_snaps_up_infiniband_technology_product_line_from_qlogic/. 
  26. ↑ "Oracle Engineers Its Own InfiniBand Interconnects". http://www.nextplatform.com/2016/02/22/oracle-engineers-its-own-infiniband-interconnects/. 
  27. ↑ https://nvidianews.nvidia.com/news/nvidia-to-acquire-mellanox-for-6-9-billion

External links



Retrieved from "https://handwiki.org/wiki/index.php?title=InfiniBand&oldid=27815"