AMD EPYC Processors come to Oracle Cloud Infrastructure for HPC

Print Friendly, PDF & Email

AMD has announced the availability of the first AMD EPYC processor-based instance on Oracle Cloud Infrastructure.

The AMD EPYC processor ‘E’ series will lead with the bare metal, Standard ‘E2’, available immediately as the first instance type within the Series. At $0.03/Core hour, the AMD EPYC instance is up to 66 per cent less on average per core than general purpose instances offered by the competition.

With the launch of the AMD instance, Oracle has once again demonstrated that we are focused on getting the best value and performance to our customers,’ said Clay Magouyrk, senior vice president, software development, Oracle Cloud Infrastructure. ‘At greater than 269 GB/Sec, the AMD EPYC platform, offers the highest memory bandwidth of any public cloud instance. Combined with increased performance, these cost advantages help customers maximize their IT dollars as they make the move to the cloud.”

In addition to the bare metal offering, today’s release also features the one, two, four and eight core VM Shapes. The new offering takes advantage of the AMD’s core count, memory bandwidth, I/O capability, as well as advanced security features that come with the AMD EPYC processor.

We are delighted that Oracle is adding AMD EPYC processors to its cloud offerings. The EPYC processor provides more cores, more memory bandwidth, and outstanding stability. That translates into leadership TCO for cloud deployments. It is a perfect fit for Oracle customers running their business applications in the cloud,” said Forrest Norrod, senior vice president and general manager, Datacenter and Embedded Solutions Business Group, AMD. “Our work with Oracle highlights how the EPYC processors’ unique design offers cloud users a significant price/performance advantage.”

With 64 cores per server and up to 33 per cent more memory channels than comparable x86 instances, the Standard E2 instance is suited for data analytics workloads that demand higher cores and memory bandwidth.

Within the Hadoop ecosystem, AMD has partnerships with many of the leading providers including Cloudera, Hortonworks, MapR and Transwarp. On a complete 10TB Terasort run the AMD instance demonstrated up to 40 per cent savings in cost per Terasort as compared to other x86 instances.

For HPC workloads the higher memory bandwidth of the EPYC instance fits customer use cases such as weather modeling, computational fluid dynamics, simulation and crash analysis in aviation and automotive manufacturing, and oil and gas exploration.

These instances are generally available in Oracle Cloud Infrastructure’s US East-Ashburn region today, London by the end of October and will be available in other US and European regions by the end of the year.

This story appears here as part of a cross-publishing agreement with Scientific Computing World.

Sign up for our insideHPC Newsletter