Xilinx and Mipsology Release Integrated Solution for High-Performance Inference in Data Center
Mipsology’s Zebra Accelerates Deep Learning Inference on Xilinx’ Alveo Boards
SAN JOSE, CALIF. , and PALAISEAU, FRANCE –– November 13, 2018 –– Xilinx, Inc. (NASDAQ: XLNX) and Mipsology, an innovative neural network acceleration startup, today announced that Mipsology’s Zebra neural network accelerator computes neural networks on Xilinx’ Alveo U200 at speeds of 2,000 images per second for ResNet50, and 3,700 images per second on Xilinx’ Alveo U250.
“Processing neural networks with Zebra on Xilinx’ Alveo gives users a faster solution than traditional GPUs at a lower price than CPUs, making FPGAs ideal devices for deep learning computation,” states Ludo Larzul, Mipsology’s founder and chief executing officer.
Combining Xilinx’s Alveo and Mipsology’s Zebra replaces CPUs and GPUs for neural network inference. Xilinx Alveo boards deliver high-performance computation in data center, while Zebra’s ease of use and high throughput enable Xilinx FPGAs of various sizes to compute any convolutional neural network with zero effort. The Alveo/Zebra solution offers 4x faster acceleration than the most popular GPU board when classifying images based on ResNet50 neural network using TensorFlow as framework.
“Zebra caught our attention with the level of acceleration achievable on our FPGAs, specifically on our Alveo cards,” says Ramine Roane, Xilinx’s Vice President of marketing. “Combining speed with the ability to deliver a pre-programmed accelerator gives Zebra a clear advantage over all other CPU or GPU acceleration solutions.”
Xilinx and Mipsology will exhibit at SC18 now through Thursday, November 15. Xilinx will be in Booth #2018. Mipsology will demonstrate Zebra in Booth #3871 as part of the Startup Pavilion.
About Xilinx
Xilinx develops highly flexible and adaptive processing platforms that enable rapid innovation across a variety of technologies – from the endpoint to the edge to the cloud. Xilinx is the inventor of the FPGA, hardware programmable SoCs and the ACAP, designed to deliver the most dynamic processor technology in the industry and enable the adaptable, intelligent and connected world of the future. For more information, visit www.xilinx.com
About Mipsology
Mipsology is a startup developing state-of-the-art FPGA-based accelerators targeted for deep learning applications in neural networks. It was founded in 2015 by a team of engineers and scientists who created a family of world-class FPGA-based super-computers over the past 20 years. More information is available at: www.mipsology.com
|
Related News
- Rambus Unveils PCIe 7.0 IP Portfolio for High-Performance Data Center and AI SoCs
- Xilinx Versal HBM Series with Integrated High Bandwidth Memory Tackles Big Data Compute Challenges in the Network and Cloud
- Xilinx Extends Data Center Leadership with New Alveo U280 HBM2 Accelerator Card; Dell EMC First to Qualify Alveo U200
- Xilinx Launches the World's Fastest Data Center and AI Accelerator Cards
- IAR Systems enables high-performance machine learning based on latest neural network library from Arm
Breaking News
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Cadence Unveils Arm-Based System Chiplet
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |