Super MicroComputer, Inc. (Nasdaq: SMCI), a global leader in high-performance computing, storage, networking solutions and green computing technology, today announced two new systems that support Intel’s Arctic Sound-M and Intel Habana Labs Gaudi®2.
The systems allow customers to take advantage of the latest acceleration technology from Intel and Intel Habana, which are ideal for cloud gaming, media delivery, artificial intelligence (AI) and machine learning (ML) workloads.
“Supermicro continues to work closely with Intel and Habana Labs to deliver a range of server solutions that support Arctic Sound-M and Gaudi2. This enables us to deliver solutions to organizations in need of efficient media delivery and AI training, “said Charles Liang, President and CEO.” We continue to work with leading technology vendors to deliver total solutions for complex workloads while improving system performance. ”
Thanks to its Building Block Solutions approach, Supermicro can quickly bring new technologies to market. The method makes it easy to fit new GPUs and acceleration technology into existing designs or, if necessary, quickly adapt an existing design as needed for better performing components.
“Supermicro helps deliver advanced AI and media processing with systems utilizing our latest Gaudi2 and Arctic Sound-M accelerators,” said Sandra Rivera, executive vice president and general manager of Datacenter and AI Group at Intel. “Supermicros Gaudi AI Training Server will accelerate deep learning training.”
Supermicro systems with Arctic Sound-M GPUs meet the high demands of cloud gaming, media transcoding and streaming, virtual desktop infrastructure (VDI), simulation and visualization, machine learning and content creation. With the industry’s first hardware AV-1 encoder and open source media software stack, Intel Arctic Sound-M significantly improves performance compared to software video transcoding and delivery solutions. In addition, the system includes acceleration functions for VDI environments.
The Intel Arctic Sound-M GPUs will initially be available in 2U 2 Node single processor Intel system with 3 GPUs per. node, the 4U 10xGPU system and the CloudDC server, with additional systems to be announced later this year. The new Supermicro AI training servers include dual Intel 3rd generation Intel Xeon scalable processors and eight Habana Gaudi2 accelerators for high-performance AI training environments.
Habana Labs Gaudi2
Habana Labs Gaudi2 excels when it comes to workloads in various applications, including: vision applications such as image classification, object detection and NLP (Natural Language Processing) models. The new AI Training server will be the first commercial rollout with the new Habana Gaudi2 (HL-225) in an 8U chassis. This server accelerates AI training to new levels of performance when paired with dual 3rd generation Intel Xeon scalable processors and up to 8TB of DRAM. With 24 hot-swappable drive bays that support ample local high-performance storage, significant amounts of IO are found on the server. In addition, the Gaudi2 accelerator is available with air cooling (HL-225). With Habana Gaudi2, scaling is simple and straightforward. Thanks to ROCE, each Habana Gaudi2 accelerator can communicate with the other Gaudi2 accelerators at 700 GB / s. In addition, each server includes 6 x QSFP-DD ports for easy scaling to handle larger models and datasets.