Skip to main content

Supermicro's New Multi-Node Liquid Cooled Architecture with Maximum Performance Density Purpose-Built for HPC at Scale

The New Rack-Scale Ready FlexTwin™ Systems Deliver Unprecedented Compute-Density in a Multi-Node Form Factor, with DLC dual CPUs up to 500W Each per Node, Front Node Access and Optimized Networking

San Jose, Calif. – September 19, 2024 – Supermicro, Inc. (NASDAQ: SMCI) a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge is announcing the all-new FlexTwin family of systems which has been designed to address the needs of scientists, researchers, governments, and enterprises undertaking the world’s most complex and demanding computing tasks. Featuring flexible support for the latest CPU, memory, storage, power and cooling technologies, FlexTwin is purpose-built to support demanding HPC workloads including financial services, scientific research, and complex modeling. These systems are cost-optimized for performance per dollar and can be customized to suit specific HPC applications and customer requirements thanks to Supermicro’s modular Building Block Solutions® design.

"Supermicro's FlexTwin servers set a new standard of performance density for rack-scale deployments with up to 96 dual processor compute nodes in a standard 48U rack," said Charles Liang, president and CEO of Supermicro. "At Supermicro, we're able to offer a complete one-stop solution that includes servers, racks, networking, liquid cooling components, and liquid cooling towers, speeding up the time to deployment and resulting in higher quality and reliability across the entire infrastructure, enabling customers faster time to results. Up to 90% of the server generated heat is removed with the liquid cooling solution, saving significant amounts of energy and enabling higher compute performance."

For more information about FlexTwin systems, please visit here.

This new multi-node design incorporates Supermicro's modular Resource Saving Architecture which uses shared power supplies and DLC of critical components to reduce raw materials usage, maximize power efficiency, and lower data center PUE (Power Usage Effectiveness). The new FlexTwin architecture includes a range of new and industry standard technologies which not only improve performance but also enhance workload flexibility and serviceability for large-scale data centers.

  • Support for the latest generation of high-frequency CPUs up to 500W with DLC enabling compute densities unachievable with traditional data center air cooling
  • Multi-vendor CPU support with up to 12 memory channels per CPU
  • Front-accessible hot-swap compute nodes, I/O ports, and optional drive bays to enhance serviceability and simplify maintenance from the cold aisle
  • Enhanced reliability with redundant power supplies and hot-swappable liquid cooling pumps to minimize downtime
  • Optimized total rack level solutions for in-row and in-rack liquid cool deployments
FlexTwin

To support the deployment of the FlexTwin architecture at scale, Supermicro offers rack-scale integration services to design, build, validate, and deliver complete solutions of any size thanks to an industry-leading global manufacturing capacity of up to 5,000 racks per month (including 1,350 liquid cooled racks), extensive rack-scale integration and testing facilities, and a comprehensive suite of management software solutions.