Dolphin Interconnect Solutions

Last updated
Dolphin Interconnect Solutions
OSE: DOLP
Founded1989;35 years ago (1989)
Headquarters Oslo, Norway
Website www.dolphinics.com

Dolphin Interconnect Solutions is a privately held manufacturer of high-speed data communication systems headquartered in Oslo, Norway [1] [2] and Woodsville, New Hampshire, USA.

Contents

The technology of Dolphin was based on development work at Norsk Data during the late 1980s. Dolphin Interconnect Solutions was founded in 1992 as a spin-off from Dolphin Server Technology which was, in turn, a spin-off from Norsk Data in 1989. [3] [4] Dolphin Interconnect Solutions develops technology for low latency and high-speed communication between servers and/or embedded computer systems.

History

Dolphin Server Technology emerged from Norsk Data, "a formerly flourishing Norwegian minicomputer maker", with one of its aims to build a business developing systems based on the Motorola 88000 architecture, these being adopted by Norsk Data as the new company's initial customer, with the intention of gradually reducing Norsk Data's stake to less than 50 percent and thus gradually increasing the new company's independence. [5] Dolphin was established as a consequence of the restructuring of Norsk Data in 1988, the separation of product development from other aspects of the parent company's business, and the transfer of 125 employees from Norsk Data's development divisions. Another initial activity was to develop a Unix system for Norsk Data's existing ND-5000 range. [6] : 6 

Following an initial product announcement in late 1989, [7] by April 1990, Dolphin Server Technology had started shipping products in its Triton 88 series, based on the Motorola 88000 processor family, with these systems supporting up to four processors. Compliant with the 88open Consortium's standards, the Triton 88 series ran a Unix product developed by UniSoft, providing binary compatibility with contemporary 88000-based systems. Dolphin offered these products through value-added resellers in European, North American, and South American markets, also cultivating business with original equipment manufacturers, resulting in the Triton 88 models appearing "under several different brand names" worldwide. [8]

Having announced plans for an emitter-coupled logic (ECL) version of the Motorola 88000, projected to run at 125 MHz, executing up to eight instructions in parallel, and delivering a peak performance of 1000 MIPS, [9] Dolphin Server Technology participated in the development and standardization of Scalable Coherent Interface (SCI) technology, delivering the first prototype in 1992 for an implementation of the base SCI standard as a gate array fabricated by Vitesse Semiconductor. A CMOS implementation was demonstrated in 1994 in association with LSI Logic. [10]

They were among the pioneers in the development and commercialization of the Scalable Coherent Interface (SCI) technology. SCI was one of the earliest high-speed interconnect technologies designed to enable multiprocessing and data sharing across a wide area, significantly enhancing the performance of clustered computing environments. This innovation helped set the stage for the development of modern high-performance computing networks.

The company had announced its plans for the ECL variant of the 88000, named Orion and developed in conjunction with Motorola, in December 1989. This processor, employing a technique called "instruction folding" originating from research done within Norsk Data, involved "a mutual exchange of patented technology" between the companies. It was hoped that Orion would ship in the first half of 1992. [8] Initially mentioned in Norsk Data's 1988 annual report, Orion was also the subject of a project under the auspices of the EUREKA programme in association with Siemens and NTNU, "building on four different technologies from Siemens AG, Motorola Inc., National Semiconductor Ltd. and Dolphin". [6] : 36 

In 1993, Dolphin, described as a vendor of "RISC-based UNIX multiprocessor servers" specializing in solutions for the government and banking, announced a deal with NeXT to resell NeXT computer products and to license NeXT's software technology. [11] Ultimately, Dolphin "abandoned the server market entirely". [4] : 472 

Products

Dolphin started out continuing the development of a line of SCI products [12] from Norsk Data, by implementing customer-specific technology, as well as providing Peripheral Component Interconnect (PCI) and later PCIe boards for commodity and high-performance computing (HPC) systems. Sun Microsystems agreed to re-sell Dolphin's SCI interfaces for the SBus in 1996. [13] Dolphin SCI products are available under the Dolphin Express SCI label.

The StarFabric product line was added through the acquisition of StarGen Inc. in early 2007. [14] StarGen shareholders received about 22% of the combined company. StarGen, which became the US subsidiary of Dolphin, had been based in Marlborough, Massachusetts, and led by Tim Miller. [15] StarFabric provides a PCI-based interconnect running over standard Ethernet Category 5 cables. Similarly, the Dolphin Express DX product line introduced in 2006 was also acquired from StarGen. DX was based on the Advanced Switching Interconnect (ASI) standard and implements a PCI Express Gen1 switched-topology technology supporting both host to host communication and host to IO expansion over an individual cable connection. [16] A company called Numascale was spun out of Dolphin in 2008, referring to the concept of non-uniform memory access (NUMA). [17]

The Dolphin Express IX product line introduced in 2010 is based on PCIe Gen2 and Gen3 integrated circuits ("chips") from Integrated Device Technology. The Dolphin Express PX product line, introduced in 2016, is based on PCI Express Gen3 chipsets from Broadcom. IX and PX implements a PCIe-native switched-topology technology.

The SISCI application programming interface (API) was developed for the shared memory SCI hardware. With the introduction of the PX, IX, and DX-line of products, the API was expanded to support features like reflective memory, [18] multicast and PCIe peer-to-peer communication.

SuperSockets is a software platform for Dolphin Express providing a low latency, high throughput implementation of the Berkeley sockets and Winsock APIs. [19] It was introduced in 2007. [20]

SmartIO has been used to share GPUs, NVMe drives, and other devices in a PCIe network. [21] [22] [23] [24]

Related Research Articles

<span class="mw-page-title-main">Non-uniform memory access</span> Computer memory design used in multiprocessing

Non-uniform memory access (NUMA) is a computer memory design used in multiprocessing, where the memory access time depends on the memory location relative to the processor. Under NUMA, a processor can access its own local memory faster than non-local memory. NUMA is beneficial for workloads with high memory locality of reference and low lock contention, because a processor may operate on a subset of memory mostly or entirely within its own cache node, reducing traffic on the memory bus..

<span class="mw-page-title-main">OpenStep</span> Defunct object-oriented application programming interface specification

OpenStep is an object-oriented application programming interface (API) specification developed by NeXT. It provides a framework for building graphical user interfaces (GUIs) and developing software applications. OpenStep was designed to be platform-independent, allowing developers to write code that could run on multiple operating systems, including NeXTSTEP, Windows NT, and various Unix-based systems. It has influenced the development of other GUI frameworks, such as Cocoa for macOS, and GNUstep.

The 88000 is a RISC instruction set architecture developed by Motorola during the 1980s. The MC88100 arrived on the market in 1988, some two years after the competing SPARC and MIPS. Due to the late start and extensive delays releasing the second-generation MC88110, the m88k achieved very limited success outside of the MVME platform and embedded controller environments. When Motorola joined the AIM alliance in 1991 to develop the PowerPC, further development of the 88000 ended.

<span class="mw-page-title-main">Expansion card</span> Circuit board for connecting to a computer system to add functionality

In computing, an expansion card is a printed circuit board that can be inserted into an electrical connector, or expansion slot on a computer's motherboard to add functionality to a computer system. Sometimes the design of the computer's case and motherboard involves placing most of these slots onto a separate, removable card. Typically such cards are referred to as a riser card in part because they project upward from the board and allow expansion cards to be placed above and parallel to the motherboard.

Originally, the word computing was synonymous with counting and calculating, and the science and technology of mathematical calculations. Today, "computing" means using computers and other computing machines. It includes their operation and usage, the electrical processes carried out within the computing hardware itself, and the theoretical concepts governing them.

HyperTransport (HT), formerly known as Lightning Data Transport, is a technology for interconnection of computer processors. It is a bidirectional serial/parallel high-bandwidth, low-latency point-to-point link that was introduced on April 2, 2001. The HyperTransport Consortium is in charge of promoting and developing HyperTransport technology.

<span class="mw-page-title-main">InfiniBand</span> Network standard

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. Between 2014 and June 2016, it was the most commonly used interconnect in the TOP500 list of supercomputers.

<span class="mw-page-title-main">PCI Express</span> Computer expansion bus standard

PCI Express, officially abbreviated as PCIe or PCI-e, is a high-speed serial computer expansion bus standard, designed to replace the older PCI, PCI-X and AGP bus standards. It is the common motherboard interface for personal computers' graphics cards, capture cards, sound cards, hard disk drive host adapters, SSDs, Wi-Fi, and Ethernet hardware connections. PCIe has numerous improvements over the older standards, including higher maximum system bus throughput, lower I/O pin count and smaller physical footprint, better performance scaling for bus devices, a more detailed error detection and reporting mechanism, and native hot-swap functionality. More recent revisions of the PCIe standard provide hardware support for I/O virtualization.

<span class="mw-page-title-main">Scalable Coherent Interface</span> High-speed interconnect standard for shared memory multiprocessing and message passing

The Scalable Coherent Interface or Scalable Coherent Interconnect (SCI), is a high-speed interconnect standard for shared memory multiprocessing and message passing. The goal was to scale well, provide system-wide memory coherence and a simple interface; i.e. a standard to replace existing buses in multiprocessor systems with one with no inherent scalability and performance limitations.

<span class="mw-page-title-main">Norsk Data</span> Defunct Norwegian computer manufacturer

Norsk Data was a minicomputer manufacturer located in Oslo, Norway. Existing from 1967 to 1998, it had its most active period from the early 1970s to the late 1980s. At the company's peak in 1987, it was the second largest company in Norway and employed over 4,500 people.

<span class="mw-page-title-main">Quadrics (company)</span>

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.

<span class="mw-page-title-main">Chipset</span> Electronic component to manage data flow of a CPU

In a computer system, a chipset is a set of electronic components on one or more integrated circuits that manages the data flow between the processor, memory and peripherals. The chipset is usually found on the motherboard of computers. Chipsets are usually designed to work with a specific family of microprocessors. Because it controls communications between the processor and external devices, the chipset plays a crucial role in determining system performance.

Aviion was a series of computers from Data General that were the company's main product from the late 1980s until the company's server products were discontinued in 2001. Earlier Aviion models used the Motorola 88000 CPU, but later models moved to an all-Intel solution when Motorola stopped work on the 88000 in the early 1990s. Some versions of these later Intel-based machines ran Windows NT, while higher-end machines ran the company's flavor of Unix, DG/UX.

<span class="mw-page-title-main">PCI-X</span> Computer bus and expansion card standard

PCI-X, short for Peripheral Component Interconnect eXtended, is a computer bus and expansion card standard that enhances the 32-bit PCI local bus for higher bandwidth demanded mostly by servers and workstations. It uses a modified protocol to support higher clock speeds, but is otherwise similar in electrical implementation. PCI-X 2.0 added speeds up to 533 MHz, with a reduction in electrical signal levels.

<span class="mw-page-title-main">Blade server</span> Server computer that uses less energy and space than a conventional server

A blade server is a stripped-down server computer with a modular design optimized to minimize the use of physical space and energy. Blade servers have many components removed to save space, minimize power consumption and other considerations, while still having all the functional components to be considered a computer. Unlike a rack-mount server, a blade server fits inside a blade enclosure, which can hold multiple blade servers, providing services such as power, cooling, networking, various interconnects and management. Together, blades and the blade enclosure form a blade system, which may itself be rack-mounted. Different blade providers have differing principles regarding what to include in the blade itself, and in the blade system as a whole.

<span class="mw-page-title-main">Altix</span> Supercomputer family

Altix is a line of server computers and supercomputers produced by Silicon Graphics, based on Intel processors. It succeeded the MIPS/IRIX-based Origin 3000 servers.

<span class="mw-page-title-main">RapidIO</span> High-speed interconnect technology

The RapidIO architecture is a high-performance packet-switched electrical connection technology. It supports messaging, read/write and cache coherency semantics. Based on industry-standard electrical specifications such as those for Ethernet, RapidIO can be used as a chip-to-chip, board-to-board, and chassis-to-chassis interconnect.

<span class="mw-page-title-main">Motorola 88110</span> Microprocessor developed by Motorola

The MC88110 was a microprocessor developed by Motorola that implemented the 88000 instruction set architecture (ISA). The MC88110 was a second-generation implementation of the 88000 ISA, succeeding the MC88100. It was designed for use in personal computers and workstations.

Coherent Accelerator Processor Interface (CAPI), is a high-speed processor expansion bus standard for use in large data center computers, initially designed to be layered on top of PCI Express, for directly connecting central processing units (CPUs) to external accelerators like graphics processing units (GPUs), ASICs, FPGAs or fast storage. It offers low latency, high speed, direct memory access connectivity between devices of different instruction set architectures.

Compute Express Link (CXL) is an open standard for high-speed, high capacity central processing unit (CPU)-to-device and CPU-to-memory connections, designed for high performance data center computers. CXL is built on the serial PCI Express (PCIe) physical and electrical interface and includes PCIe-based block input/output protocol (CXL.io) and new cache-coherent protocols for accessing system memory (CXL.cache) and device memory (CXL.mem). The serial communication and pooling capabilities allows CXL memory to overcome performance and socket packaging limitations of common DIMM memory when implementing high storage capacities.

References

  1. ↑ Wagner, N., Tiffany, P., Peterson, S. D. (2013). Business Plans For Canadians For Dummies. United States: Wiley.
  2. ↑ Tveito, A., Bruaset, A. M., Lysne, O. (2009). Simula Research Laboratory: By Thinking Constantly about it. Germany: Springer Berlin Heidelberg.
  3. ↑ "Dolphin Interconnect is on its Way". Tech Monitor. Retrieved October 25, 2021.
  4. 1 2 Gustavson, David B. (1999). SCI - Scalable coherent interface : architecture and software for high-performance compute clusters. Springer. p. 472. ISBN   3-540-66696-6. ISSN   0302-9743 . Retrieved 24 February 2023.
  5. ↑ Faden, Michael (February 1990). "Industry Report: RISC on the Rise in Europe". UNIX Review. pp. 14, 16, 19, 21, 23, 25–26. Retrieved 24 February 2023.
  6. 1 2 Annual Report 1988 (PDF). Norsk Data. May 1989. Retrieved 7 July 2024.
  7. ↑ "Nordic Nereus". UNIX Review. December 1989. p. 117. Retrieved 26 February 2023.
  8. 1 2 Wilson, David (July 1991). "Tested Mettle: The Dolphin Server Technology Triton 88". UNIX Review. pp. 38, 40, 43–44, 46, 48. Retrieved 24 February 2023.
  9. ↑ "Microbytes". Byte. February 1990. p. 17. Retrieved 24 February 2023.
  10. ↑ Lundstrom, Stephen F. (1994). Defining the Global Information Infrastructure: Infrastructure, Systems, and Services. SPIE–The International Society for Optical Engineering. pp. 157–158. ISBN   0-8194-1680-0.
  11. ↑ Bergson, Eliot (January 1993). "NeXT hooks into European server". NeXTWORLD. p. 4. Retrieved 24 February 2023.
  12. ↑ Liaaen M.C., Kohmann H. (1999) Dolphin SCI Adapter Cards. In: Hellwagner H., Reinefeld A. (eds) SCI: Scalable Coherent Interface. Lecture Notes in Computer Science, vol 1734. Springer, Berlin, Heidelberg. https://doi.org/10.1007/10704208_5
  13. ↑ "Sun Microsystems and Dolphin Sign OEM Agreement". EE Times. October 10, 1996. Retrieved October 25, 2021.
  14. ↑ Rick Merritt (January 26, 2007). "Dolphin, Stargen merge to ride Express". EE Times. Retrieved October 25, 2021.
  15. ↑ Rick Merritt (February 12, 2007). "Merger to yield I/O, clustering products based on PCI Express". EE Times. Retrieved October 25, 2021.
  16. ↑ Venkata Krishnan; Tim Miller; Herman Paraison (September 17, 2007). "Dolphin express: A transparent approach to enhancing PCI Express". 2007 IEEE International Conference on Cluster Computing. pp. 464–467. doi:10.1109/CLUSTR.2007.4629266. ISBN   978-1-4244-1387-4. S2CID   14004892.
  17. ↑ Timothy Prickett Morgan (November 24, 2010). "Numascale brings big iron SMP to the masses". EThe Register. Retrieved October 25, 2021.
  18. ↑ "Dolphin Launches Low Latency Networking Solution". Wall Street & Technology.
  19. ↑ Ryan, Stein Jørgen; Bryhni, Haakon (March 30 – April 3, 1998). "Eliminating the protocol stack for socket based communication in shared memory interconnects". In José Rolim (ed.). Lecture Notes in Computer Science. 10th International Parallel and Distributed Processing, held in conjunction with the 12th International Parallel Processing Symposium and 9th Symposium on Parallel and Distributed Processing. Vol. 1388. Orlando, Florida, USA: Springer. doi:10.1007/3-540-64359-1_724.
  20. ↑ Rick Merritt (April 16, 2007). "Dolphin, Stargen Swims After InfiniBand in Clusters". EE Times. Retrieved October 25, 2021.
  21. ↑ "GTC Silicon Valley-2019: Dynamic Sharing of GPUs and IO in a PCIe Network". NVIDIA Developer. April 28, 2019.
  22. ↑ Markussen, Jonas; Kristiansen, Lars Bjørlykke; Halvorsen, Pål; Kielland-Gyrud, Halvor; Stensland, Håkon Kvale; Griwodz, Carsten (July 8, 2021). "SmartIO: Zero-overhead Device Sharing through PCIe Networking". ACM Transactions on Computer Systems. 38 (1–2): 2:1–2:78. doi: 10.1145/3462545 . hdl: 10852/86684 .
  23. ↑ "PCIe with low cost and low latency, ideal for sharing resources". www.microcontrollertips.com.
  24. ↑ "Automotive Industry Eyes Low-Cost, Low-Latency PCIe for Sharing Resources – Lynnette Reese".
This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.