 | This article needs additional citations for verification .(September 2014) |
Myrinet, ANSI/VITA 26-1998, is a high-speed local area networking system designed by the company Myricom to be used as an interconnect between multiple machines to form computer clusters.
Myrinet was promoted as having lower protocol overhead than standards such as Ethernet, and therefore better throughput, less interference, and lower latency while using the host CPU. Although it can be used as a traditional networking system, Myrinet is often used directly by programs that "know" about it, thereby bypassing a call into the operating system.
Myrinet physically consists of two fibre optic cables, upstream and downstream, connected to the host computers with a single connector. Machines are connected via low-overhead routers and switches, as opposed to connecting one machine directly to another. Myrinet includes a number of fault-tolerance features, mostly backed by the switches. These include flow control, error control, and "heartbeat" monitoring on every link. The "fourth-generation" Myrinet, called Myri-10G, supported a 10 Gbit/s data rate and can use 10 Gigabit Ethernet on PHY, the physical layer (cables, connectors, distances, signaling). Myri-10G started shipping at the end of 2005.
Myrinet was approved in 1998 by the American National Standards Institute for use on the VMEbus as ANSI/VITA 26-1998. [1] One of the earliest publications on Myrinet is a 1995 IEEE article. [2]
| Generation | Year [3] | Bandwidth | Notes | |
|---|---|---|---|---|
| 1 | Myrinet | 1994 | 0.64 Gbit/s | PECL PHY |
| 2 | Myrinet-LAN | 1996 | 1.28 Gbit/s | Backwards compatible with original Myrinet. LVDS PHY for 1.28 Gbit/s operation. [4] |
| Myrinet-SAN | System-Area Network, shorter reach (3m) but lower cost [5] | |||
| 3 | Myrinet-2000 | 2000 | 2 Gbit/s | Backwards-compatible with Gigabit Ethernet |
| 4 | Myri-10G | 2005 | 10 Gbit/s | Backwards-compatible with 10 Gigabit Ethernet |
Myrinet is a lightweight protocol with little overhead that allows it to operate with throughput close to the basic signaling speed of the physical layer. For supercomputing, the low latency of Myrinet is even more important than its throughput performance, since, according to Amdahl's law, a high-performance parallel system tends to be bottlenecked by its slowest sequential process, which in all but the most embarrassingly parallel supercomputer workloads is often the latency of message transmission across the network.
According to Myricom, 141 (28.2%) of the June 2005 TOP500 supercomputers used Myrinet technology. In the November 2005 TOP500, the number of supercomputers using Myrinet was down to 101 computers, or 20.2%, in November 2006, 79 (15.8%), and by November 2007, 18 (3.6%), a long way behind gigabit Ethernet at 54% and InfiniBand at 24.2%.
In the June 2014 TOP500 list, the number of supercomputers using Myrinet interconnect was 1 (0.2%). [6] [7]
In November, 2013, the assets of Myricom (including the Myrinet technology) were acquired by CSP Inc. [8] In 2016, it was reported that Google had also offered to buy the company. [9]
In computer architecture, a bus is a communication system that transfers data between components inside a computer, or between computers. This expression covers all related hardware components and software, including communication protocols.
Ethernet is a family of wired computer networking technologies commonly used in local area networks (LAN), metropolitan area networks (MAN) and wide area networks (WAN). It was commercially introduced in 1980 and first standardized in 1983 as IEEE 802.3. Ethernet has since been refined to support higher bit rates, a greater number of nodes, and longer link distances, but retains much backward compatibility. Over time, Ethernet has largely replaced competing wired LAN technologies such as Token Ring, FDDI and ARCNET.
A supercomputer is a computer with a high level of performance as compared to a general-purpose computer. The performance of a supercomputer is commonly measured in floating-point operations per second (FLOPS) instead of million instructions per second (MIPS). Since 2017, supercomputers have existed which can perform over 1017 FLOPS (a hundred quadrillion FLOPS, 100 petaFLOPS or 100 PFLOPS). For comparison, a desktop computer has performance in the range of hundreds of gigaFLOPS (1011) to tens of teraFLOPS (1013). Since November 2017, all of the world's fastest 500 supercomputers run on Linux-based operating systems. Additional research is being conducted in the United States, the European Union, Taiwan, Japan, and China to build faster, more powerful and technologically superior exascale supercomputers.
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. By 2014, it was the most commonly used interconnect in the TOP500 list of supercomputers, until about 2016.
HIPPI, short for High Performance Parallel Interface, is a computer bus for the attachment of high speed storage devices to supercomputers, in a point-to-point link. It was popular in the late 1980s and into the mid-to-late 1990s, but has since been replaced by ever-faster standard interfaces like Fibre Channel and 10 Gigabit Ethernet.
VMEbus is a computer bus standard physically based on Eurocard sizes.
Futurebus, or IEEE 896, is a computer bus standard, intended to replace all local bus connections in a computer, including the CPU, memory, plug-in cards and even, to some extent, LAN links between machines. The effort started in 1979 and didn't complete until 1987, and then immediately went into a redesign that lasted until 1994. By this point, implementation of a chip-set based on the standard lacked industry leadership. It has seen little real-world use, although custom implementations continue to be designed and used throughout industry.
A network interface controller is a computer hardware component that connects a computer to a computer network.
Quadrics was a supercomputer company formed in 1996 as a joint venture between Alenia Spazio and the technical team from Meiko Scientific. They produced hardware and software for clustering commodity computer systems into massively parallel systems. Their highpoint was in June 2003 when six out of the ten fastest supercomputers in the world were based on Quadrics' interconnect. They officially closed on June 29, 2009.
TCP offload engine (TOE) is a technology used in some network interface cards (NIC) to offload processing of the entire TCP/IP stack to the network controller. It is primarily used with high-speed network interfaces, such as gigabit Ethernet and 10 Gigabit Ethernet, where processing overhead of the network stack becomes significant. TOEs are often used as a way to reduce the overhead associated with Internet Protocol (IP) storage protocols such as iSCSI and Network File System (NFS).
In computing, remote direct memory access (RDMA) is a direct memory access from the memory of one computer into that of another without involving either one's operating system. This permits high-throughput, low-latency networking, which is especially useful in massively parallel computer clusters.
In computer networking, jumbo frames are Ethernet frames with more than 1500 bytes of payload, the limit set by the IEEE 802.3 standard. The payload limit for jumbo frames is variable: while 9000 bytes is the most commonly used limit, smaller and larger limits exist. Many Gigabit Ethernet switches and Gigabit Ethernet network interface controllers and some Fast Ethernet switches and Fast Ethernet network interface cards can support jumbo frames.
The IBM BladeCenter was IBM's blade server architecture, until it was replaced by Flex System in 2012. The x86 division was later sold to Lenovo in 2014.
The Texas Advanced Computing Center (TACC) at the University of Texas at Austin, United States, is an advanced computing research center that is based on comprehensive advanced computing resources and supports services to researchers in Texas and across the U.S. The mission of TACC is to enable discoveries that advance science and society through the application of advanced computing technologies. Specializing in high performance computing, scientific visualization, data analysis & storage systems, software, research & development and portal interfaces, TACC deploys and operates advanced computational infrastructure to enable the research activities of faculty, staff, and students of UT Austin. TACC also provides consulting, technical documentation, and training to support researchers who use these resources. TACC staff members conduct research and development in applications and algorithms, computing systems design/architecture, and programming tools and environments.
A system on a module (SoM) is a board-level circuit that integrates a system function in a single module. It may integrate digital and analog functions on a single board. A typical application is in the area of embedded systems. Unlike a single-board computer, a SoM serves a special function like a system on a chip (SoC). The devices integrated in the SoM typically requires a high level of interconnection for reasons such as speed, timing, bus width etc.. There are benefits in building a SoM, as for SoC; one notable result is to reduce the cost of the base board or the main PCB. Two other major advantages of SoMs are design-reuse and that they can be integrated into many embedded computer applications.
Magerit is the name of the one of the most powerful supercomputers in Spain. It also reached the second best Spanish position in the TOP500 list of supercomputers. This computer is installed in CeSViMa, a research center of the Technical University of Madrid.
QPACE is a massively parallel and scalable supercomputer designed for applications in lattice quantum chromodynamics.
10 Gigabit Ethernet is a group of computer networking technologies for transmitting Ethernet frames at a rate of 10 gigabits per second. It was first defined by the IEEE 802.3ae-2002 standard. Unlike previous Ethernet standards, 10GbE defines only full-duplex point-to-point links which are generally connected by network switches; shared-medium CSMA/CD operation has not been carried over from the previous generations of Ethernet standards so half-duplex operation and repeater hubs do not exist in 10GbE. The first standard for faster 100 Gigabit Ethernet links was approved in 2010.
QPACE 2 is a massively parallel and scalable supercomputer. It was designed for applications in lattice quantum chromodynamics but is also suitable for a wider range of applications..