Arteris’ Post

#DellTechnologies has worked with http://Scalers.ai and Broadcom to develop a PoC for detecting Pneumonia in X-rays. Using a single Dell #PowerEdge server with AMD CPUs, an AI model was trained and ran on the CPU alone, and pneumonia could be detected in X-rays with significant accuracy. Watch video: http://tdas.so/D9A4B2

Density and perf/ watts of AMD Turin vs ARM 192c shows Great results and benefits with full x86 , current and long term benefits right now. Sustainabilty now and for long term

#DellTechnologies has worked with http://Scalers.ai and Broadcom to develop a PoC for detecting Pneumonia in X-rays. Using a single Dell #PowerEdge server with AMD CPUs, an AI model was trained and ran on the CPU alone, and pneumonia could be detected in X-rays with significant accuracy. Watch video: http://tdas.so/9EF194

#DellTechnologies has worked with http://Scalers.ai and Broadcom to develop a PoC for detecting Pneumonia in X-rays. Using a single Dell #PowerEdge server with AMD CPUs, an AI model was trained and ran on the CPU alone, and pneumonia could be detected in X-rays with significant accuracy. Watch video: http://tdas.so/32D019

How does AMD EPYC Turin 192C stack up vs AmpereOne 192C? And BTW, many 'Turin' systems launched by OEMs. Summary: EPYC 1.6x performance at only 1.2x average power. EPYC Turin significantly lower idle power. AMD EPYC 9965 "Turin Dense" Delivers Better Performance/Power Efficiency vs. AmpereOne 192-Core ARM CPU https://lnkd.in/gue2x-_i    EPYC 9965 average power ~275 Watts AmpereOne A192-32X average power ~230 Watts EPYC 9965 idle power ~19 watts AmpereOne A192-32X idle power ~101 Watts   "Turin Dense dispels the myth that x86_64 can't be as power efficient as ARM or having cores as a small." Note: only a subset of Phoronix were used due to ability to run on ARM. #amd #epyc #turin

What's better, large cache or HBM? Which one speeds up workloads the most? Of course, SOC architecture matters a lot as well (aka a big parking lot with one lane in and out is not necessarily a good parking lot). "Every cache miss that I can eliminate leads to a small performance increase by both decreasing the memory system load (and leaving more headroom for the workload), and by eliminating a CPU stall on DRAM access." The test: 32-core: The 32-core AMD EPYC 9384X versus the 32-core Intel® Xeon® CPU Max 9462 in both High Bandwidth Memory (HBM) and Cache modes. Top of stack: The 96-core AMD EPYC 9684X versus the 56-core Intel Xeon CPU Max 9480 in both HBM and Cache modes. #amd #epyc #togetherweadvance #genoa Raghu Nambiar Read more at: https://lnkd.in/g5iBZkyF

I've often said, if your going to accelerate, accelerate a LOT, not a little. AMD EPYC + MaxLinear accelerate A LOT. No need for complex sku selection or scalability concerns, no noisy neighbors stealing CPU memory bandwidth, no forcing max core count CPUs, no waiting on CPU design cycle to get new innovations.......list goes on.... Installing MaxLinear Panther III accelerators in systems powered by 4th Gen AMD EPYC processors delivers outstanding performance for workloads that require data compression, deduplication, and encryption/decryption backed by optimal power usage, a mature and proven SW stack, and superior RAS compared to a comparable Intel QAT system. Read the white paper: https://lnkd.in/gs_sNCgm Luke Chang Mark Moran Kishore Seendripu Dan McNamara #amd #epyc #maxlinear

$249.00 $ EKWB EK-Quantum Velocity2 D-RGB For Intel Socket LGA 1700 CPU , Nickel +Acetal https://lnkd.in/gyv-_Vh4 Features: Details EK-Quantum Velocity² D-RGB – 1700 Nickel + Acetal is the new high-performance premium quality CPU water block for the new Intel® socket LGA 1700-based platform. It features a next-generation cooling engine that is socket-specific. This CPU water block is EK-Matrix7 compliant and is mounted via a patent-pending EK Exact Mount system. OptiFlow Cooling Engine EK-Quantum Velocity² series CPU water blocks embed the next generation cooling engine named OptiFlow. They use a specific combination of mounting pressure and coldplate geometry tailored for the IHS and die layout of Intel LGA 1700 socket processors. Low hydraulic flow restriction enables these […] #

Benchmarking Turnium v7.0 to push bandwidth and speed to multi-Gig. Performance testing with Intel's cutting-edge CPU technologies. 🚀 Explore the business possibilities of ultra-fast connectivity. https://bit.ly/49aQL72 #NetworkBenchmark #IntelCPUs #TurniumInnovation #SDWAN

Introducing the S7X - the ultimate 1U workhorse built for any workload. Powered by Intel Xeon 6, the S7X delivers outstanding performance and flexibility with key features such as: - Dual CPUs with up to 144 cores each - 32 x DDR5 RDIMM up to 6,400 MHz - 16 x 8,000 MHz Multi-Ranked Buffered DIMMs doubling DDR5 memory bandwidth - NEBS compliance for Telco/5G deployment Flexible Storage Options: - (12) 2.5” hot-plug NVMe/SATA/SAS drives - (16) E1.S hot-plug NVMe drives - (20) E3.S 1T hot-plug NVMe drives - (4) E3.S 1T + (8) E3.S 2T CXL devices - (2) OCP 3.0 SFF slots supporting up to 400Gb network cards Ideal for: #CloudComputing | #EnterpriseStorage | #AI #MachineLearning HYPERSCALERS #ServerInnovation #S7X