NEUCHIPS Announces World's First Deep Learning Recommendation Model (DLRM) Accelerator: RecAccel

SAN JOSE, Calif., May 12, 2020 — (PRNewswire) —  Today, NEUCHIPS Corp., an AI compute company specializing in domain-specific accelerator solutions, announced the world's first recommendation engine - RecAccelTM - that can perform 500,000 inferences per second. Running open-source PyTorch DLRM, RecAccelTM outperforms server-class CPU and inference GPU by 28X and 65X, respectively. It is equipped with an ultra-high-capacity, high-bandwidth memory subsystem for embedding table lookup and a massively parallel compute FPGA for neural network inference. Via a PCIe Gen3 host interface, RecAccelTM is ready for data center adaptation.

RecAccelTM boosts DLRM inference performance through the following innovations:

  • Embedding-specific memory architecture, allocation and access scheme.
  • Application-specific processing pipeline.
  • Scalable Multiply-And-Accumulator (MAC) array.

Most e-commerce, on-line advertisement, and internet service providers employ recommendation systems to expand interests or delivery services, such as search result ranking, friend suggestions, movie recommendations and purchase suggestions. Recommendations usually account for most of the AI inference workload in data centers. 

"Fast and accurate recommendation inference is the key to e-commerce business success," said Dr. Youn-Long Lin, CEO of NEUCHIPS. "RecAccelTM powers your business with the lowest latency, highest throughput, and best TCO."

About NEUCHIPS: NEUCHIPS Corp. is an application-specific compute solution provider based in Hsinchu, Taiwan. Founded by a team of veteran IC design experts in 2019, NEUCHIPS's mission is "Smarten AI computing through innovated IC design to make Intelligence Everywhere." NEUCHIPS management and R&D team has decades of experience in top IC design houses and holds 22 patents in signal processing, neural network, and circuits. As an OCP community member, NEUCHIPS devotes itself to provide the most cost-effective AI inference accelerators for best TCO (Total Cost of Ownership).

For more information, please visit us at www.neuchips.ai.

Press Contact:
Kinny Chen
BP Manager
NEUCHIPS Corporation
Phone: +886-3-5720720 ext. 6666
contact@neuchips.ai

Related Images

neuchips-corp.jpg
NEUCHIPS Corp.
NEUCHIPS Corp. is an application-specific compute solution provider.

Cision View original content to download multimedia: http://www.prnewswire.com/news-releases/neuchips-announces-worlds-first-deep-learning-recommendation-model-dlrm-accelerator-recaccel-301057319.html

SOURCE NEUCHIPS Corp.

Contact:
Company Name: NEUCHIPS Corp.
Web: http://www.neuchips.ai

Featured Video
Editorial
Jobs
Equipment Engineer, Raxium for Google at Fremont, California
Manufacturing Test Engineer for Google at Prague, Czechia, Czech Republic
Mechanical Test Engineer, Platforms Infrastructure for Google at Mountain View, California
Mechanical Engineer 2 for Lam Research at Fremont, California
Mechanical Engineer 3 for Lam Research at Fremont, California
Mechanical Manufacturing Engineering Manager for Google at Sunnyvale, California
Upcoming Events
Celebrate Manufacturing Excellence at Anaheim Convention Center Anaheim CA - Feb 4 - 6, 2025
3DEXPERIENCE World 2025 at George R. Brown Convention Center Houston TX - Feb 23 - 26, 2025
TIMTOS 2025 at Nangang Exhibition Center Hall 1 & 2 (TaiNEX 1 & 2) TWTC Hall Taipei Taiwan - Mar 3 - 8, 2025
Automate 2025 at Detroit, Michigan, USA MI - May 12 - 15, 2025



© 2024 Internet Business Systems, Inc.
670 Aberdeen Way, Milpitas, CA 95035
+1 (408) 882-6554 — Contact Us, or visit our other sites:
AECCafe - Architectural Design and Engineering EDACafe - Electronic Design Automation GISCafe - Geographical Information Services TechJobsCafe - Technical Jobs and Resumes ShareCG - Share Computer Graphic (CG) Animation, 3D Art and 3D Models
  Privacy PolicyAdvertise