AMD, DoiT to help Google Cloud customers optimize for Epyc

Pair promise to provide the tools to achieve peak Zen

AMD has partnered with DoiT International on a program aimed at helping Google Cloud customers optimize their off-prem workloads to obtain the best performance, cost and security, at least for workloads running on AMD silicon.

DoiT and AMD today said they aim to provide Google Cloud customers with the tools, technology and expertise to ensure their compute workloads are finely tuned for performance, cost and longevity. The program will also involve technical consulting to help customers identify the right workloads for a benchmark comparison on AMD’s Epyc server chips.

DoiT sells itself as a global multicloud software and managed services provider, which means that it offers products aimed at simplifying and automating cloud operations. In particular, the firm claims that its FlexSave solution automates the management of Google Cloud CUDs, or committed use discounts, which provide users with discounted prices in exchange for a commitment to use a minimum level of resources for a specified term.

As part of the new program, Google Cloud customers will be offered access to workload performance benchmark analysis, technical consulting and DoiT’s suite of solutions, including the FlexSave offering, all free of charge for DoiT customers, according to AMD.

“We are excited to collaborate with DoiT to allow more customers to test and optimize on the latest Google Cloud technology based on AMD Epyc processors, improving performance and maximizing their cloud adoption costs,” AMD corporate vice president for Cloud Business Lynn Comp said in a statement.

Google Cloud, like others, has been adopting AMD’s third-generation Epyc server processors (code-named Milan) for their combination of performance and price since they became available last year. Last week, it declared availability of new compute optimized (C2D) virtual machines running on AMD’s silicon, configurable with up to 112 vCPUs (56 cores), 896GB of memory and up to 3TB of local SSD storage. C2D instances are available today in regions around the globe.

Other cloud companies are also moving to offer instances based on the Zen-based Epyc family. This week, AWS announced availability of its own compute-optimized Amazon EC2 C6a instances, which the firm claims offer a 15 per cent improvement in price-performance versus C5a instances, and cost 10 per cent less than comparable x86-based EC2 instances.

Amazon’s C6a instances are available in 10 sizes, ranging from 2 up to 192 vCPUs and up to 384GB of memory. The new instances are available in the US East, US West and EU AWS Regions. ®