Kategoria: NVIDIA H100 accelerators
Superkomputer Intel Aurora przełamuje barierę eksaskali, stając się drugim najpotężniejszym superkomputerem
—
in AI-accelerated high performance computing (HPC), AI-driven research, Alps machine, AMD, AMD EPYC CPUs, AMD Instinct MI250X accelerators, Argonne Leadership Computing Facility, Argonne National Laboratory, Asia, Aurora, Aurora Early Science Program (ESP), cancer treatments, CERN’s Large Hadron Collider Project (LHC), China, Cray ClusterStor E1000 parallel storage platform, Cray’s Slingshot 11 network, dark universe, Energy and Sustainable Economic Development’s (ENEA) CRESCO 8, Euro-Mediterranean Centre on Climate Change’s (CMCC) Cassandra, EuroHPC/CSC, Europe, exaFLOPS, exascale machine, Falcon Shores, Frontier, fuel-efficient aircraft, Fugaku, GREEN500, Hewlett Packard Enterprise (HPE), High Performance Computing, High-Performance Conjugate Gradient (HPCG), HPE Cray EX architecture, HPE Cray Shasta supercomputer architecture, HPE Slingshot interconnect, HPL-MxP benchmark, IBM processors, Intel, Intel Data Center GPU Max Series accelerators, Intel Tiber™ Developer Cloud, Intel Xe GPU architecture, Intel Xeon CPU Max series processors, Intel® architecture, Intel® Data Center GPU Max Series, Intel® Distributed Asynchronous Object Storage (DAOS) technology, Intel® Gaudi® 2, Intel® Kubernetes Service, Intel® oneAPI DPC++/C++ Compiler, Intel® Optane™ Persistent Memory, Intel® Xeon® 6 E-core and P-core systems, Intel® Xeon® 6 with P-cores and Multiplexer Combined Ranks (MCR) memory., Intel® Xeon® CPU Max Series, ISC High Performance 2024, Italian National Agency for New Technologies, Linux Foundation’s Unified Acceleration Foundation (UXL), LUMI system, Microsoft Azure Cloud, neuroscience research, North America, NVIDIA H100 accelerators, Oak Ridge National Laboratory, oneAPI programming model, Slingshot-11 interconnect, Swiss National Supercomputing Centre, Texas Advanced Computing Center (TACC), TOP500, United Kingdom Atomic Energy Authority (UKAEA), United States[ad_1] System Frontier w Oak Ridge National Laboratory w Tennessee pozostaje najpotężniejszym superkomputerem, z wynikiem HPL wynoszącym 1,206 EFlop/s. System Aurora w Argonne Leadership Computing Facility w Illinois jest teraz drugą maszyną, która oficjalnie przekroczyła barierę eksaskali z wynikiem HPL wynoszącym 1,012 EFlop/s. System Eagle w chmurze Microsoft Azure jest najwyżej sklasyfikowanym systemem chmurowym na…
