Name: Chinese Exascale Supercomputer Prototype Tested with AI Workloads
Brand: tomshardware.com
SKU: tomshardware-com-19
Availability: OnlineOnly

critic

Chinese Exascale Supercomputer Prototype Tested with AI Workloads

preview

Tom's Hardware is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. .Chinese scientists have started to test a prototype of locally developed exascale supercomputer using convolutional neural networks (CNN) AI workloads. As reported by&nbsp;&nbsp;the Tianhe-3 machine runs&nbsp;Armv8-based Phytium 2000+ (FTP) and the Matrix 2000+ (MTP) processors that were primarily designed for traditional supercomputing with full FP64 precision.&nbsp;&nbsp;Supercomputing and artificial intelligence are poised to become important competitive advantages that technologically developed companies and countries have against others in the coming years. Exascale supercomputers will be instrumental for traditional simulations, data analytics, and scientific discoveries, whereas AI supercomputers will be used for various pathfinding tasks, such as drug discovery or space exploration that do not necessarily require FP64 precision. The key thing is that to be competitive, supercomputing sites will need to support both AI and HPC workloads, which is why modern supercomputers use both CPUs and GPUs to have maximum flexibility and to support all kinds of capabilities.&nbsp;Being behind their American rivals, Chinese developers are exploring multiple approaches to exascale systems, including 'traditional' CPU + GPU architecture and&nbsp;&nbsp;with vector units. Yet another approach tested by Chinese researchers is Tianhe-3 machine that uses proprietary&nbsp;Armv8-based Phytium 2000+ (FTP) and the Matrix 2000+ (MTP) processor/node architecture, reports&nbsp;.&nbsp;The Phytium-designed FTP part has 64 cores distributed over eight eight-core 'panels' each featuring its own memory controller unit (MCU), 2MB L2 cache per every four cores, and 32KB L1 cache per core. The 64-core FTP operates at 2.20 GHz and delivers 614.4 FP64 GFLOPS performance. To put the number into contest, Fujitsu's 48-core&nbsp;&nbsp;for the&nbsp;&nbsp;has a&nbsp;theoretical peak performance of 3.3792&nbsp;FP64 TFLOPS. Meanwhile, the&nbsp;MTP has four&nbsp;'super nodes' featuring 32 cores and two MCUs per node. The report does not disclose power consumption of FTP and MTP, it also omits peak performance of the latter.To test the supercomputing architecture, the researchers used a CNN AI workload with a single sample input, but found that the node is "weak and the processor performance itself cannot be fully exerted." Meanwhile, CNN performance with multiple inputs was found to be considerably better as there was a "huge improvement of both two processes (GEMM and Conv2D with the increase in input samples." Still, while FTP could approach its peak performance, MTP could not be fully exploited because of low memory bandwidth.&nbsp;&nbsp;"We evaluated the training performance of CNN on the Tianhe-3 prototype," the abstract description of the&nbsp;. "The performance of image convolution and matrix multiplication on the FTP and MTP was tested to evaluate the single-node performance, and the Allreduce element was tested to evaluate the scalability of the distributed training on the prototype cluster. We also qualitatively analyzed the performance bottlenecks of CNN on the FTP and MTP processors by Roofline model and provided some optimization suggestions for improving the CNN on the Tianhe-3 prototype."&nbsp;It is noteworthy that unlike upcoming exascale supercomputers developed in the U.S., Tianhe-3 was not optimized for AI workloads, so the very fact that it can run them is significant. Meanwhile, peak FP64 performance of the&nbsp;Phytium 2000+does not seem to be very high when compared to Fujitsu's A64FX, so it remains to be seen whether it is reasonable to use such a CPU for exascale systems. Furthermore, with Phytium blacklisted by the U.S. government, it is unclear if the CPU developer can use a more advanced fabrication process to scale its Arm architecture for supercomputers to gain performance and lower power consumption.Get instant access to breaking news, in-depth reviews and helpful tips.Thank you for signing up to Tom's Hardware. You will receive a verification email shortly.There was a problem. Please refresh the page and try again.MOST POPULARMOST SHARED1234512345

Tom's Hardware is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. .Chinese scientists have started to test a prototype of locally developed exascale supercomputer using convolutional neural networks (CNN) AI workloads. As reported by&nbsp;&nbsp;the Tianhe-3 machine runs&nbsp;Armv8-based Phytium 2000+ (FTP) and the Matrix 2000+ (MTP) processors that were primarily designed for traditional supercomputing with full FP64 precision.&nbsp;&nbsp;Supercomputing and artificial intelligence are poised to become important competitive...

Chinese Exascale Supercomputer Prototype Tested with AI Workloads

No article scores yet.

CRITIC REVIEWS

PUBLIC REVIEWS