Centaur Releases In-Depth Analysis from The Linley Group for World’s First x86 Processor with AI Coprocessor Technology
Detailed Technical Disclosure Follows Successful Demonstration at ISC EastTrade Show
Austin, Texas –December 9, 2019 – Centaur Technology today revealed in-depth information aboutits new processor-design technology for integrating high-performance x86 CPUs with a specialized coprocessor optimized for artificial intelligence (AI) acceleration. On its website(www.centtech.com), Centaur provides a new independentreportfromThe Linley Group, the industry’s leading authority on microprocessor technology and publishers of Microprocessor Report.The Linley Group reviewed Centaur’s detailed design documents andinterviewedCentaur’s CPU and AI architectstosupport the analysis of both Centaur’s newest x86 microarchitecture and the AI coprocessordesign.
“Centaur is galloping back into the x86 market with an innovative processor design thatcombines eight high-performance CPUs with a custom deep-learning accelerator (DLA). Thecompany is the first to announce a server-processor design that integrates a DLA. The newaccelerator, called Ncore, delivers better neural-network performance than even the mostpowerful Xeon, but without the high cost of an external GPU card,” statedLinley Gwennap, Editor-in-Chief, Microprocessor Report.
The Linley Group referenced certified MLPerf1benchmark (Preview) scoresto compare Centaur’s AI performance to high-end x86 CPUcoresfrom theleadingx86 vendor. Based on MLPerfscores, Centaur’s AI-coprocessorinference performanceiscomparable to 23of Intel’s world-class x86 cores that now support512-bit vector neural network instructions (VNNI)2.Centaur’s AI coprocessor uses an architecturallysimilar single-instruction-multiple-data (SIMD) approachas VNNI, but crunches 32,768bitsin a single clock cycleusing a 16MB memory with 20 terabytes/sec of bandwidth. Moreover, by offloading inference processing to a specialized coprocessor, the x86 CPU cores remain available for other, more general-purpose tasks. Application developers can innovate new algorithms that take advantage of the unparalleled inference latency enabled by Centaur’s AI performance and tight integration with x86 CPUs.
Attendees at the ISC East trade show in NYCsaw Centaur’s new technology up closefor the first time. The demoshowcasedvideo analyticsusing Centaur’s reference systemwithx86-basednetwork-video-recording (NVR) software from Qvis Labs. In addition to conventional,real-time object detection/classification, Centaurwas the only vendor at the show tohighlight leading-edge applicationssuch as semantic segmentation (pixel-level image classification) and a new technique for human pose estimation(“stickfigures”).Centaurisfocusedon improving the hardware price/performance and software productivity for platforms to support this next wave ofresearch applications andspeed deployment into new server-class products.About Centaur Technology Austin, Texas-based Centaur Technology is a small group of very talented engineers that have been designing AI accelerator technology for high-performance, low-cost x86-compatible microprocessors. Over the past 24 years, Centaur Technology has shipped 26 different x86-based designs with millions of units sold. More information is at www.centtech.com.
[1] MLPerf v0.5 Inference Closed/Preview audited submission, Sept. 2019. MLPerf name and logo are trademarks. See www.mlperf.org for more information.
[2] MLPerf Inf‑0.5–23. Dual Intel® Xeon® Platinum 9282 (112 total cores).MobileNet-V1 and ResNet-50 on Intel x86coreswith VNNI. Intel and Xeon are trademarks of Intel Corporation.