Supermicro launches 12 Xeon 6+ servers with up to 576 E-cores per system

Supermicro has introduced 12 new server platforms built for Intel Xeon 6+ processors, targeting high-density cloud, virtualization, 5G analytics, content delivery, and other throughput-heavy workloads. The company’s pitch is improved performance-per-watt and higher core density for data center deployments.

The systems are based on what Supermicro describes as its X14 platforms and are designed to support Xeon 6+ configurations with up to 288 efficiency cores (E-cores) per socket. In dual-socket servers, that scales to up to 576 E-cores per server. Supermicro also points to the Xeon 6+ platform’s generational claims of higher throughput, including “double the core count,” up to 17% higher instructions per clock (IPC), five times more last-level cache, and 25% faster memory support versus prior generations.

Supermicro grouped the 12 systems into four product families. The Hyper series covers single- and dual-socket 1U and 2U rackmount systems, with support for high-memory configurations and “advanced networking.” SuperBlade uses a blade chassis format that fits up to 10 compute nodes in a 6U chassis to increase rack compute density with shared infrastructure. FlexTwin is positioned as a high-density, liquid-cooled platform where each dual-socket node operates independently while sharing power and cooling resources. And GrandTwin is a single-socket, multi-node design aimed at high core-count configurations and E-core-heavy workloads, with an emphasis on density and thermal efficiency.

For data center engineers, the big takeaway is that E-core-heavy CPU platforms are pushing operators to re-check familiar constraints: rack power budgets, cooling headroom, and per-rack network uplinks can all become the limiting factor before “CPU capacity” does. Dense 1U/2U rackmount designs and multi-node chassis can amplify those tradeoffs, especially when the goal is squeezing more cores into the same footprint.

“By working closely with Intel, we have optimized our DCBBS with the new Xeon 6+ processors to deliver breakthrough core density and efficiency,” said Charles Liang, president and CEO of Supermicro. Supermicro also said its DCBBS approach is a modular infrastructure offering built from “validated components and subsystems,” scaling from individual servers and networking to rack-scale and data center-level solutions, including software and services.

Source: Supermicro

Get Data Center Engineering News In Your Inbox:

Popular Posts:

picotest-thumbnail
A closer look at power integrity at AI scale
Not-All-Liquids-Are-Created-Equal-White-Paper-FINAL-1
Download the practical guide to liquid cooling fluid selection
John Deboer - Siemens - Designing for 100X - when speed wins, reliability starts to give
Designing for 100X: when speed wins, reliability starts to give
Schneider-Electric-launches-Uniflair-XCA-oil-free-chillers-up-to-2
Schneider Electric launches Uniflair XCA oil-free chillers up to 2.5 MW for AI data centers
Leak Testing Liquid-Cooled Server Racks Poppe+Potthoff
Testing the weak spots in liquid-cooled server racks: design the connection like it’s the product

Share Your Data Center Engineering News

Do you have a new product announcement, webinar, whitepaper, or article topic? 

Get Data Center Engineering News In Your Inbox: