Jump to Content

Expanding the Tau VM family with Arm-based processors

July 13, 2022
Subra Chandramouli

Product Manager

Jamie Kinney

Product Manager

Organizations that are developing ever larger, scale-out applications will leave no stone unturned in their search for a compute platform that meets their needs. For some, that means looking to the Arm® architecture. Known for delivering excellent performance per watt efficiency, Arm-based chips are already ubiquitous in mobile devices, and have proven themselves for supercomputing workloads. At Google Cloud, we’re also excited about using Arm chips for the next generation of scale-out, cloud-native workloads.

Last year, we added Tau VMs to Compute Engine, offering a new family of VMs optimized for cost-effective performance for scale-out workloads. Today we are thrilled to announce the Preview release of our first VM family based on the Arm architecture, Tau T2A. Powered by Ampere® Altra® Arm-based processors, T2A VMs deliver exceptional single-threaded performance at a compelling price. Tau T2A VMs come in multiple predefined VM shapes, with up to 48 vCPUs per VM, and 4GB of memory per vCPU. They offer up to 32 Gbps networking bandwidth and a wide range of network attached storage options, making Tau T2A VMs suitable for scale-out workloads including web servers, containerized microservices, data-logging processing, media transcoding, and Java applications.

Google Cloud customers and developers now have the option of choosing an Arm-based Google Cloud VM to test, develop and run their workloads on the optimal architecture for their workload. Several of our customers have had private preview access to T2A VMs for the last few months and have had a great experience with these new VMs. Below is what few of them have to say about T2A VMs.

“Our drug discovery research at Harvard includes several compute-intensive workloads that run on SLURM using VirtualFlow1. The ability to run our workloads on tens of thousands of VMs in parallel is critical to optimize compute time. We ported our workload to the new T2A VM family from Google and were up and running with minimal effort. The improved price-performance of the T2A will help us screen more compounds and therefore discover more promising drug candidates.” - Christoph Gorgulla, Research Associate, Harvard University

“In recent years, we have come to rely on Arm-based servers to power our engineering activity at lower cost and higher performance compared to legacy environments. The introduction of the Arm Neoverse N1-based T2A instance allows us to diversify our use of cloud compute on Arm-based hardware and leverage the Google Compute Engine to build the exact virtual machine types we need, with the convenience of Google Kubernetes Engine for containerized workloads.” – Mark Galbraith, Vice President, Productivity Engineering, Arm.

Ampere Computing and Arm have been key partners for Google Cloud in delivering this VM.

“Ampere® Altra® Cloud Native Processors were designed from the ground up to meet the demands of modern cloud applications,” said Jeff Wittich, Chief Product Officer, Ampere Computing. “Our close collaboration with Google Cloud has resulted in the launch of the new price-performance optimized Tau T2A instances, which enable demanding scale-out applications to be deployed rapidly and efficiently."

"Arm Neoverse was developed to support the complex demands of the modern cloud and cloud-native applications and enable cloud operators to achieve their ambitious sustainability goals,” said Chris Bergey, Senior Vice President and General Manager, Infrastructure Line of Business, Arm. “Built upon Arm's unmatched performance efficiency, Google Cloud’s Arm-based VMs provide even more choice when running scale-out workloads.”

Integration with Google Cloud services 

Google Cloud is ramping up its support for Arm. T2A VMs support most popular Linux operating systems such as RHEL, Suse Linux Enterprise Server, CentOS, Ubuntu, and Rocky Linux. In addition, T2A VMs also support Container-optimized OS to bring up Docker containers quickly, efficiently and securely. Further, developers building applications on Google Cloud can already use several Google Cloud services with T2A VMs — with more coming later this year:

  • Google Kubernetes Engine - Google Kubernetes Engine (GKE) is the leading platform for organizations looking for advanced container orchestration. Starting today, GKE customers can run their containerized workloads using the Arm architecture on T2A. Arm nodes come packed with key GKE features, including the ability to run in GKE Autopilot mode for a hands-off experience. Read more about running your Arm workloads with GKE here

  • Batch - Our newly launched Batch service supports T2A. As of today users will be able to run batch jobs on T2A instances to optimize their cost of running workloads.

  • Dataflow - Dataflow is a fully managed streaming analytics service that minimizes latency, processing time, and cost through autoscaling and batch processing. You can now use T2A VMs with your Dataflow workloads.

Extensive ISV partner ecosystem

While Arm chips are relative newcomers to data center workloads, there’s already a robust ecosystem of ISV support for Tau T2A VMs. In fact, Ampere lists more than 100 applications, databases, cloud-native software and programming languages that are already running on Ampere-based T2A VMs, with more being added all the time. Further, ISV partners that have validated their solutions on T2A VMs have been impressed by the ease with which they were able to port their software to Tau T2A VMs. 

“Momento’s serverless cache enables developers to accelerate database and application performance at scale. Over the past few months, we have become intimately familiar with Google Cloud’s new T2A VMs. We were pleasantly surprised with the ease of portability to Arm instance from day one. The maturity of the T2A platform gives us the confidence to start using these VMs in production. Innovations like T2A VMs in Google Cloud help us continuously innovate on behalf of our customers.” - Khawaja Shams, CEO, Momento. Learn more about Momento’s T2A experience.

“SchedMD’s Slurm open-source workload manager is designed specifically to satisfy the demanding needs of compute-intensive workloads. We are thrilled with the introduction of the T2A VMs on Compute Engine. The introduction of T2A will give our customers more choice of virtual machines for their demanding workload management needs using Slurm.” - Nick Ihli, Director of cloud and Solutions Engineering, SchedMD.

"At Rescale, we help our customers deliver innovations faster with high performance computing built for the cloud. We are excited to now offer T2A VMs to our customers, with compelling price-performance to further drive engineering and scientific breakthroughs. With Arm-based VMs on Google Cloud, we are able to offer our customers a larger portfolio of solutions for computational discovery.” - Joris Poort, CEO, Rescale. Learn more about Rescale’s integrations with T2A VMs.

“Canonical Ubuntu is a popular choice for developers seeking a third party server operating system running on Google Cloud, and we are very happy to provide Ubuntu as the guest OS for users of Compute Engine on Google Cloud's new Arm-based VMs, which supports our most recent long-term supported versions. Once migrated, users will find a completely familiar environment with all the packages and libraries they know and rely on to manage their workloads.” - Alexander Gallagher, VP of Cloud Sales at Canonical. Learn more about Canonical’s ​​optimized Ubuntu image.

To help you get started, we’re providing customers, ISV and ecosystem partners access to T2A VMs at no charge for a trial period, to help jumpstart development on Ampere Arm-based processors. When Tau T2A reaches General Availability later this year, we’ll continue to offer a generous trial program that offers up to 8 vCPUs and 32 GB of RAM at no cost. (updated on Oct 5th 2022: From October 5, 2022 to April 5, 2023 Tau T2A VMs are available for a 6-month free trial. Monthly credit of up to $222 per billing account to use T2A VM resources will be available for Google Cloud users in the us-central1 region. You can create multiple VMs, however once the total cost of your resources exceeds $222, your account is billed for any additional resources you are using. This offer only covers T2A VM costs. Other Compute Engine resources and services are charged to your billing account.).

Pricing and availability

Tau T2A VMs are price-performance optimized for your cloud-native applications. A 32vCPU VM with 128GB RAM will be priced at $1.232 per hour for on-demand usage in us-central1. 

T2A VMs are generally available (updated on October 5th 2022) in several Google Cloud regions: us-central (Iowa - Zone A,B,F), europe-west4 (Netherlands - Zone A,B, C) and asia-southeast1 (Singapore - Zone B,C) and will be in General Availability in the coming months. We look forward to working with you as you explore using Ampere Arm-based T2A VMs for your next scale-out workload in the cloud.

To learn more about Tau T2A VMs or other Compute Engine VM options, check out our machine types and pricing pages. To get started, go to the Google Cloud Console and select T2A for your VMs.

1. https://www.nature.com/articles/s41586-020-2117-z
Posted in