2X more All-Flash NVMe Bandwidth, Up to 12TB* Memory, Open OCP 3.0 I/O, Maximum GPU Acceleration, Robust 5G, as well as Power Savings, and the Highest Available Compute Density
Super Micro Computer, Inc. (Nasdaq: SMCI), a global leader in enterprise computing, storage, networking, and green computing technology, introduces the industry’s most complete server lineup incorporating the latest 3rd Gen Intel Xeon Scalable processors delivering leading performance and reduced TCO and Total Cost to the Environment (TCE). The new X12 systems are already deployed in volume at innovative organizations such as Osaka University and DISH.
“Supermicro is delivering the most performant energy saving systems, pushing the envelope with new systems incorporating the latest Intel processors,” said Charles Liang, president and CEO, Supermicro. “Our new Hyper product line is designed for maximum performance and can house 32 DIMMs of high-speed memory, four high-performance GPUs, and up to 400 Gbit LAN, which supports our customers’ most demanding requirements. We work closely with technology partners across the industry to deliver a broad range of servers and storage systems optimized for many workloads and delivering the best solutions, from the highest performance systems, like Hyper or our Delta GPU deep learning optimized platform, to cost-efficient TCO optimized platforms, such as our new 2U 2-Node video streaming platform. We are proud to design and manufacture these systems at our Silicon Valley headquarters and our rapidly expanding Taiwan facilities.”
Most Comprehensive Range of Systems with Innovative Technology
The entire line of over 100 application-optimized systems, including Hyper, SuperBlade®, the Twin Product Family (BigTwin ®, TwinPro ®, and FatTwin ®), Ultra, CloudDC, GPU, Telco/5Gand Edge servers, feature the 3rd Gen Intel Xeon Scalable processors. The top-of-the-line Intel Xeon Platinum 8380 processor with 40 cores running at 2.3 GHz in the Ultra and Hyper platforms delivers leadership performance. The network-optimized CPUs are leveraged in a comprehensive portfolio of 5G and embedded edge systems. The Big Twin multi-node and SuperBlade systems with Intel Speed Select CPUs deliver flexibility and efficiency for Cloud Service Providers. Also, the single-processor WIO and storage systems bring compelling cost savings for general purpose and storage applications. The portfolio includes water cooled configurations for high power CPUs in high-density systems in HPC installations with a direct liquid cooling option to remove the limitations of traditional air-cooling for heat dissipation and thus lowers overall data center operating costs.
“Our 3rd Gen Intel Xeon Scalable processors have been designed to deliver true flexibility. They are the only data center processors with built in AI acceleration and are built to serve all workloads from edge to cloud. Our partnership with Supermicro enables innovative IT providers from every industry to deliver quality services across all of these workloads and adapt quickly while providing consistent performance,” stated Lisa Spelman, Corporate Vice President and General Manager of the Xeon and Memory Group. “We continue to work closely with Supermicro to deliver more value to customers by creating and enabling solutions that solve their greatest challenges.”
With Artificial Intelligence and Machine Learning growth, Supermicro has introduced a portfolio of new GPU systems based on 3rd Gen Intel Xeon Scalable processors. These systems leverage the 2X increase in PCI-E 4.0 I/O to deliver improved workload performance. The portfolio includes highly integrated 4U 8 GPU systems for large training workloads, and 2U 2-Node single processor GPU systems with improved power efficiency for cloud gaming and streaming. The CloudDC platform, with a balanced architecture with dedicated I/O and all-flash storage for each CPU, enables the system to deliver lower latency, higher performance, and allows remote direct memory access (RDMA) in large-scale cloud environments.
The Supermicro X12 product lineup also includes 5G edge-optimized systems, including the Ultra-E, Hyper-E, and the 2U UP 210P Edge server. The high-performance Hyper-E delivers unparalleled processing power and expansion capability while meeting demanding telco requirements, such as NEBS Level 3 certification, short system depth, and front-accessible I/O. The Hyper-E is ideal for 5G networks and edge applications that require local processing and need to include AI algorithms in a small form factor within low power and thermally restrictive environments. At the edge, Supermicro, with its broad lineup of X12 systems, can easily support a range of data collection, computing, and AI requirements.
“As we build the nation’s first cloud-based 5G network, Intel and Supermicro will help maximize the power of next-generation connectivity from the core to the edge,” said Sidd Chenumolu, DISH Vice President of Technology Development. “DISH has worked with both companies for a long time, and we are pleased to continue this great partnership.”
Supermicro has already deployed systems with leading high-performance computing research institutions, including Cybermedia Center Osaka University that is implementing the 3rd Gen Intel Xeon Scalable processor-based systems for the supercomputing system named SQUID (Supercomputer for Quest to Unsolved Interdisciplinary Data science).
“Researchers and scientists in Osaka University are now working on their own research projects in the hope of bringing open innovation to our society. We are excited to leverage the latest generation of Intel and Supermicro’s solutions for advanced scientific research projects necessitating HPC (High-Performance Computing) and HPDA (High-Performance Data Analysis).” – Dr. Susumu Date Cybermedia Center, Osaka University.
Additionally, the X12 product line allows customers to extract more value from larger datasets to take advantage of the latest Intel Optane persistent memory 200 series. The new X12 systems can access data faster, with twice the I/O performance utilizing PCI-E 4.0. With support for up to eight memory channels per socket running at 3200Mhz, more memory can be directly addressed at higher speeds, increasing application performance by up to 46% on popular benchmarks.
The new X12 systems have improved security and management capabilities. With full memory encryption, the X12 portfolio is more secure than any previous generation of Supermicro systems, giving enterprises the confidence to protect valuable customer data and deliver results securely. Supermicro’s SuperServers support Intel’s Software Guard Extensions, which provide enclaves of protected memory to run applications and their data.
The Supermicro X12 servers also have a new intelligent management system with enhanced security with hardware Root of Trust (ROT), Redfish 1.8 support, and an improved Supermicro Intelligent Management web interface. Supermicro’s cloud infrastructure management software, SuperCloud Composer, now features support for X12 platforms, bringing speed, agility, and simplicity to IT administration by integrating data center tasks into a single intelligent management solution. The SuperCloud Composer also monitors and manages the broad portfolio of multi-generation Supermicro servers and third-party systems through its data center lifecycle management feature.
Several architected solutions using the 3rd Gen Intel Xeon Scalable processors are coming soon and will be available from Supermicro, including Oracle Linux and SAP-certified solutions. Many different workloads will benefit from these new processors, including HPC, AI, Enterprise, and Cloud workloads.
* Based on 2-socket platform with Intel Optane persistent memory 200 series to be able to support 12TB per platform