Blog

We have given you the flexibility and ability to run the largest and most complex high performance computing (HPC) workloads with Amazon Elastic Compute Cloud (Amazon EC2) instances that feature enhanced networking like C5n, C6gn, R5n, M5n, and our recently launched HPC instances Hpc6a.

We heard feedback from customers asking us to deliver more options to support their most intensive workloads with higher per-vCPU compute performance as well as larger memory and local disk storage to reduce job completion time for data-intensive workloads like Finite Element Analysis (FEA) and seismic processing.

Announcing Amazon EC2 Hpc6id Instance for HPC Workloads
Today, we announce the general availability of Amazon EC2 Hpc6id instances, a new instance type that is purpose-built for tightly coupled HPC workloads. Amazon EC2 Hpc6id instances are powered by 3rd Gen Intel Xeon Scalable processors (Ice Lake) that run at frequencies up to 3.5 GHz, 1024 GiB memory, 15.2 TB local SSD disk, 200 Gbps Elastic Fabric Adapter (EFA) network bandwidth, which is 4x higher than R6i instances.

Amazon EC2 Hpc6id instances have the best per-vCPU HPC performance when compared to similar x86-based EC2 instances for data-intensive HPC workloads.

Here are the detailed specs:

Amazon EC2 Hpc6id Instances Use Cases
Customers running license-bound scenarios can lower infrastructure and HPC software licensing costs with Hpc6id. Other customers with HPC codes that are optimized for Intel-specific features, such as Math Kernel Library or AVX-512, can migrate their largest HPC workloads to Hpc6id and scale up their workloads on AWS by taking advantage of 200 Gbps EFA bandwidth.

Other customers using HPC software codes that are optimized for per-CPU performance are also able to consolidate their workloads on fewer nodes and complete jobs faster with Hpc6id. Faster job completion time helps customers to reduce both infrastructure and software licensing costs. Customers can use Hpc6id instances to quickly carry out complex calculations across a range of cluster sizes—up to tens of thousands of cores.

Customers also can use Hpc6id instances with AWS ParallelCluster to provision Hpc6id instances alongside other instance types, giving customers the flexibility to run different workload types within the same HPC cluster. Hpc6id instances benefit from the AWS Nitro System, a rich collection of building blocks that offloads many of the traditional virtualization functions to dedicated hardware and software to deliver high performance, high availability, and high security while also reducing virtualization overhead.

Now Available
Amazon EC2 Hpc6id instances are available for purchase as On-Demand or Reserved Instances or with Savings Plans. Hpc6id instances are available in the US East (Ohio) and AWS GovCloud (US-West) Regions. To optimize Amazon EC2 Hpc6id instances networking for tightly coupled workloads, use cluster placement groups within a single Availability Zone.

To learn more, visit our Hpc6 instance page and get in touch with our HPC team, AWS re:Post for EC2, or through your usual AWS Support contacts.

– Channy

Source

Originally posted on December 1, 2022 @ 12:02 pm