AWS offers instances using HPE Compute Scale-up Server 3200 : US Pioneer Global VC DIFCHQ SFO Singapore – Riyadh Swiss Our Mind

Ideal for large in-memory databases

Amazon Web Services (AWS) has launched the Elastic Compute Cloud (Amazon EC2) U7inh instance.

Part of the EC2 High Memory family, the instance has been built in collaboration with Hewlett Packard Enterprise and became generally available on December 16, according to an AWS News Blog.

HPE Compute Scale-up server 3200

HPE Compute Scale-up Server 3200– HPE

As noted by The Register, AWS does not typically name the exact servers it uses in its EC2 cloud, but has seemingly made an exception in this case.

The U7inh instance runs on the 16-socket HPE Compute Scale-up Server 3200 and is built on the AWS Nitro System.

HPE Compute Scale-up Server 3200s were launched by HPE in 2023.

The Instance has fourth-generation Intel Xeon Scalable processors, supports 32TB of memory and 1920 vCPUs. According to AWS, it offers the highest compute performance, and largest compute and memory size in AWS for running large, mission-critical database workloads.

U7inh instance offers up to two times vCPUs and 1.6 times EBS bandwidth in a single instance, compared with the largest U7i instance, a family of instances launched in May 2024.

The AWS announcement notes: “As customers continue to scale their business applications, they wanted the performance combined with the additional CPUs and memory along with SAP certification to generate real-time business insights. Other customers that currently run on-premises with HPE servers have also asked how we can help them migrate to AWS to take advantage of cloud benefits while continuing to use HPE hardware.”

This new offering may be to satisfy customer frustration that AWS servers don’t offer the same modular scaling and memory pooling that the HPE server provides.

The instances are currently available in the US East (Northern Virginia) and US West (Oregon) regions.

Other recent EC2 instance launches from AWS include UltraServers with Trainium2 chips – effectively four Trn2 instances combined into one node – and the storage-optimized I8g and I7ie instances.

https://www.datacenterdynamics.com/en/news/aws-offers-instances-using-hpe-compute-scale-up-server-3200/