LINPACK

Tachyum 8 AI Zettaflops Blueprint to Solve OpenAI Capacity Limitation

Retrieved on: 
Tuesday, December 12, 2023
Semiconductor, Data Management, Technology, Software, Artificial Intelligence, Hardware, GPU, Attention, IEEE, Economics, Organization, LMMS, OPEX, HPL, OpenAI, Ori, LINPACK, CAPEX, HPC, Azure, Microsoft Azure, Universal, AI, Microsoft, FLOPS, Prodigy, LLM, Video game, Cassette tape

The solution is ideally suited for OpenAI, others like Microsoft Azure, CoreWeave, Ori, etc.

Key Points: 
  • The solution is ideally suited for OpenAI, others like Microsoft Azure, CoreWeave, Ori, etc.
  • and research facilities needing AI datacenters that today do not have the system architecture capable of serving all interested customers.
  • For the supercomputer solution referenced in the white paper, Prodigy has a custom 46RU rack with a liquid-cooled reference design.
  • Tachyum’s software team has already integrated a software package for HPL LINPACK and other software for HPC with AI running on Prodigy FPGA soon.

Tachyum Runs Supercomputing Vector LINPACK on Prodigy FPGA

Retrieved on: 
Wednesday, December 6, 2023
Semiconductor, Data Management, Technology, Software, Artificial Intelligence, Hardware, GPU, IEEE, Economics, FPU, OPEX, HPL, Single, LINPACK, Benchmarking, Floating-point unit, Library, CAPEX, Linux, HPC, GNU Compiler Collection, Universal, AI, GCC, FMA, Prodigy, Electric motor, Video game

Tachyum ®, creator of Prodigy®, the world’s first Universal Processor, today announced it has successfully completed vector-based High-Performance LINPACK (HPL) testing using 1kb vectors on the Prodigy FPGA.

Key Points: 
  • Tachyum ®, creator of Prodigy®, the world’s first Universal Processor, today announced it has successfully completed vector-based High-Performance LINPACK (HPL) testing using 1kb vectors on the Prodigy FPGA.
  • Following scalar LINPACK benchmarks using Prodigy’s IEEE-compliant scalar Floating-Point Unit (FPU), Tachyum has now advanced to vector LINPACK.
  • Vector instructions have been verified running cleanly on the Prodigy FPGA, including vectorization in GNU Compiler Collection (GCC), libraries, Linux-supporting vectors, as well as running applications and delivering the correct results.
  • With vector LINPACK completed, Tachyum is now focused on the final stage of vector unit verification and testing with FPU: AI matrix operations.

Supermicro Offers Rack Scale Solutions with New 5th Gen Intel® Xeon® Processors Optimized for AI, Cloud Service Providers, Storage, and Edge Computing

Retrieved on: 
Thursday, December 14, 2023
Cloud, Emerald, Metaverse, Fifth, Synthetic media, SPEC, OpenFOAM, SXM, NVIDIA, DLB, EDSFF, TCO, Memory, Electricity, VDI, System, Death, BMC, Generative, InfiniBand, Graphics processing unit, CPU, Triad, Total, Communication, AI, Green computing, Enterprise, OCP, Xeon, Stock, TDP, Petascale computing, 2U, Rot, HPC, Data analysis, NIST, OAM, Data processing unit, GPU, Research, JumpStart, Compute Express Link, PCI Express, Monitoring, G1, CXL, Intel, DPU, Central processing unit, Movement, X13, LINPACK, Cloud storage, Video game, Computer data storage, Online shopping, Mobile phone, Semiconductor, Supermicro, Titanium

SAN JOSE, Calif., Dec. 14, 2023 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Manufacturer for AI, Cloud, Storage, and 5G/Edge, announces rack scale air and liquid cooled solutions based on the X13 family of workload-optimized servers now support the new 5th Gen Intel Xeon processors (formerly codenamed Emerald Rapids). The new product lineup includes GPU servers for Generative AI, throughput and latency-optimized E3.S Petascale servers, cost-effective high-density Enterprise and Simply Double storage servers for large-scale object storage, and a new 4-node SuperEdge systems with enhanced storage capacity.

Key Points: 
  • The 5th Gen Intel Xeon processors provide a 36%3 higher average performance/watt across workloads vs. 4th Gen Intel Xeon Scalable processors.
  • Both CPU and GPU direct-to-chip liquid cooling are available on the system via Supermicro's complete rack integration and liquid cooling solutions.
  • Supermicro is also launching several new servers supporting the new Intel Xeon E-2400 processors (formerly codenamed Catlow Platform, Raptor Lake-E).
  • A comprehensive software suite that enables rack management at scale for IT infrastructure solutions deployed across the core to the edge.

Supermicro Offers Rack Scale Solutions with New 5th Gen Intel® Xeon® Processors Optimized for AI, Cloud Service Providers, Storage, and Edge Computing

Retrieved on: 
Thursday, December 14, 2023
Cloud, Emerald, Metaverse, Fifth, Synthetic media, SPEC, OpenFOAM, SXM, NVIDIA, DLB, EDSFF, TCO, Memory, Electricity, VDI, System, Death, BMC, Generative, InfiniBand, Graphics processing unit, CPU, Triad, Total, Communication, AI, Green computing, Enterprise, OCP, Xeon, Stock, TDP, Petascale computing, 2U, Rot, HPC, Data analysis, NIST, OAM, Data processing unit, GPU, Research, JumpStart, Compute Express Link, PCI Express, Monitoring, G1, CXL, Intel, DPU, Central processing unit, Movement, X13, LINPACK, Cloud storage, Video game, Computer data storage, Online shopping, Mobile phone, Semiconductor, Supermicro, Titanium

SAN JOSE, Calif., Dec. 14, 2023 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Manufacturer for AI, Cloud, Storage, and 5G/Edge, announces rack scale air and liquid cooled solutions based on the X13 family of workload-optimized servers now support the new 5th Gen Intel Xeon processors (formerly codenamed Emerald Rapids). The new product lineup includes GPU servers for Generative AI, throughput and latency-optimized E3.S Petascale servers, cost-effective high-density Enterprise and Simply Double storage servers for large-scale object storage, and a new 4-node SuperEdge systems with enhanced storage capacity.

Key Points: 
  • The 5th Gen Intel Xeon processors provide a 36%3 higher average performance/watt across workloads vs. 4th Gen Intel Xeon Scalable processors.
  • Both CPU and GPU direct-to-chip liquid cooling are available on the system via Supermicro's complete rack integration and liquid cooling solutions.
  • Supermicro is also launching several new servers supporting the new Intel Xeon E-2400 processors (formerly codenamed Catlow Platform, Raptor Lake-E).
  • A comprehensive software suite that enables rack management at scale for IT infrastructure solutions deployed across the core to the edge.

UK Government Invests £225m to Create UK’s Most Powerful AI Supercomputer With University of Bristol and Hewlett Packard Enterprise

Retrieved on: 
Wednesday, November 1, 2023
Software, Hardware, Artificial Intelligence, Data Management, Technology, University, Education, Semiconductor, Government, Policy, HPE, NYSE, LINPACK, Innovation, Department of Computer Science, University of Bristol, UKRI, NCC, HPC, UK Government Decontamination Service, High Value Manufacturing Catapult, LLM, Government of the United Kingdom, Research, Hopper, Artificial intelligence, Carbon footprint, Doctorate, Hewlett Packard Enterprise, NVIDIA, Cray, Drug discovery, Carbon, Isambard, GPU, POD, Safety, AI, Robotics, Medical device, Copper, Dairy, Science, Bletchley Park, Bristol

Hewlett Packard Enterprise (HPE) will build and deliver the new system with the next generation HPE Cray EX supercomputers and over 5,000 state-of-the-art NVIDIA GH200 superchips.

Key Points: 
  • Hewlett Packard Enterprise (HPE) will build and deliver the new system with the next generation HPE Cray EX supercomputers and over 5,000 state-of-the-art NVIDIA GH200 superchips.
  • Professor Simon McIntosh-Smith, Director of the Isambard National Research Facility at the University of Bristol, said: “Isambard-AI represents a huge leap forward for AI computational power in the UK.
  • We are proud to partner with the UK Government and the University of Bristol to give UK researchers and industry access to Europe's largest AI system for open science."
  • Bristol is one of the top UK universities for AI research and scientific computing.

Supercomputer Fugaku retains first place worldwide in HPCG and Graph500 rankings

Retrieved on: 
Monday, May 22, 2023
Content as a service, Fujitsu Computer Products of America, LINPACK, Kyushu University, RIKEN, Science, HPC, Fujitsu, Graph500, AI, HPCG, Transport, Conference, Fugaku, Service

TOKYO, May 22, 2023 - (JCN Newswire) - The supercomputer Fugaku, jointly developed by RIKEN and Fujitsu, has successfully retained the top spot for seven consecutive terms in multiple major high-performance computer rankings including HPCG and Graph500 BFS (Breadth-First Search), and has also taken second place for the TOP500 and third place for the HPL-MxP (formally HPL-AI) rankings.

Key Points: 
  • TOKYO, May 22, 2023 - (JCN Newswire) - The supercomputer Fugaku, jointly developed by RIKEN and Fujitsu, has successfully retained the top spot for seven consecutive terms in multiple major high-performance computer rankings including HPCG and Graph500 BFS (Breadth-First Search), and has also taken second place for the TOP500 and third place for the HPL-MxP (formally HPL-AI) rankings.
  • The HPCG is a performance ranking for computing methods often used for real-world applications, and the Graph500 ranks systems based on graph analytic performance, an important element in data-intensive workloads.
  • The results of the rankings were announced on May 22 at ISC2023 High-Performance Computing Conference, which is currently being held in Hamburg, Germany.
  • The top ranking on Graph500 was won by a collaboration involving RIKEN, Kyushu University, Fixstars Corporation, and Fujitsu.

Fujitsu provides supercomputer system to the Japan Meteorological Agency for forecasting of linear rainbands and torrential rains

Retrieved on: 
Monday, February 27, 2023
Japan Meteorological Agency, Automated Meteorological Data Acquisition System, Forecasting, AArch64, Flood, Landslide, Meteorological Administration, System 9, Weather, Life, FUJITSU, Risk, Computer, Architecture, Observation, CPU, Arm, FLOPS, SVE, Rainband, Earthquake, Rain, Temperature, Disaster, LINPACK, Wind, RIKEN, A64FX, HBM2, Memory, Fugaku, Risk management, Cryptocurrency, Computer data storage, JMA, Fujitsu

TOKYO, Feb 27, 2023 - (JCN Newswire) - Fujitsu today announced that it provided a new supercomputer system to the Japan Meteorological Agency (hereinafter JMA) (1) for use in linear rainband forecasting.

Key Points: 
  • TOKYO, Feb 27, 2023 - (JCN Newswire) - Fujitsu today announced that it provided a new supercomputer system to the Japan Meteorological Agency (hereinafter JMA) (1) for use in linear rainband forecasting.
  • The new supercomputer system is able to predict the occurrence of linear rainbands (2) which can trigger heavy rain, leading to disasters like landslides and flooding.
  • The system is based on the "FUJITSU Supercomputer PRIMEHPC FX1000" (3) hardware, which features the same A64FX CPU as the supercomputer "Fugaku," jointly developed by RIKEN (4) and Fujitsu.
  • Reaching a theoretical peak performance of about 31.1 PFLOPS (5), the new system will help the JMA to significantly improve the accuracy of its forecasting of linear rainbands.

Settlement Reached in Lawsuit Brought by Tachyum Against Cadence

Retrieved on: 
Tuesday, January 17, 2023
Professional Services, Data Management, Semiconductor, Technology, Legal, Networks, Hardware, GPU, OPEX, Architecture, Learning, Leading edge, FPGA, HPC, FPU, Cadence Design Systems, Inc. v. Avanti Corp, Running, AI, Field-programmable gate array, Economics, LINPACK, CAPEX, SOC, Superior court, White paper, Video game console, Prodigy

Tachyum ™ today announced that it has reached an amicable settlement with Cadence Design Systems, Inc. in a lawsuit brought by Tachyum on June 8, 2022, in Santa Clara County Superior Court in California.

Key Points: 
  • Tachyum ™ today announced that it has reached an amicable settlement with Cadence Design Systems, Inc. in a lawsuit brought by Tachyum on June 8, 2022, in Santa Clara County Superior Court in California.
  • With the case now settled, Tachyum will continue its work towards reaching production-ready status of its Prodigy universal processor unabated.
  • More details about Prodigy AI features can be found in Tachyum Prodigy on the Leading Edge of AI Industry Trends White Paper PDF .
  • Tachyum has developed networking infrastructure to support HPC/AI deployments for the Prodigy Universal Processor.

Tachyum Celebrates 2022 and Announces 2023 Series C and Volume Production Plans

Retrieved on: 
Tuesday, January 3, 2023
Semiconductor, Data Management, Technology, Networks, Artificial Intelligence, Hardware, Information technology, GNU Compiler Collection, Expo, Intelligent Platform Management Interface, Faculty, GCC, Semiconductor, Economics, Prodigy (online service), Reliability, LINPACK, Czech Technical University in Prague, HPC, FPGA, UEFI, RAS, KVM, SOC, Deep learning, BMC, Availability, FPU, QEMU, Symmetric multiprocessing, Microsoft, BIOS, OPEX, Virtual machine, Shenzhen NORCO Intelligent Technology, ISC High Performance, FreeBSD, InfiniBand, QA, IEEE, Expo 2020, SMP, Security-Enhanced Linux, CAPEX, Silicon, Quality assurance, Go (programming language), Carbon, GPU, Symmetry, Computer data storage, Video game console, Prodigy, Serviceability, AI, Linux

Prodigy exceeded its performance targets and is significantly faster than any processors currently available in hyperscale, HPC and AI markets.

Key Points: 
  • Prodigy exceeded its performance targets and is significantly faster than any processors currently available in hyperscale, HPC and AI markets.
  • Tachyum completed the final piece of its core software stack with a Baseboard Management Controller (BMC) running on a Prodigy emulation system.
  • Company executives were among the featured presenters at ISC High Performance 2022 and Supercomputing 2022 events.
  • With its Series C funding, expected to close in 2023, Tachyum will finance the volume production of Prodigy Universal Processor Chip and be positioned for sustained profitability, as well as increase headcount.

Tachyum Successfully Runs LINPACK on FPGA With IEEE 754-2019 Compliant FPU

Retrieved on: 
Tuesday, November 29, 2022
Technology, Hardware, Semiconductor, Data Management, IP, UEFI, LINPACK, CAPEX, Economics, AI, GPU, RAS, FPU, Floating-point unit, HPC, OPEX, Availability, Running, Linux, Universal, EDA, Israel Inter-University Computation Center, Reliability, Prodigy, FMA, Field-programmable gate array, FPGA, IEEE, Computer data storage, Citrus, Video game console, Serviceability

This was achieved by running applications under Linux on the integer part of the processor and uses IEEE compliant Floating-Point Unit (FPU) to analyze and solve linear equations and linear least-square problems.

Key Points: 
  • This was achieved by running applications under Linux on the integer part of the processor and uses IEEE compliant Floating-Point Unit (FPU) to analyze and solve linear equations and linear least-square problems.
  • LINPACK measures a systems floating-point computing power by solving a dense system of linear equations to determine performance.
  • After successfully reaching this FPU milestone, Tachyum has only four more steps to go before the final netlist of the Prodigy processor chip.
  • Its FPU is fully IEEE compliant and corner cases have been successfully debugged.