Recently, Google announced its Cloud Virtual Machines (VMs) based on the Arm architecture of Compute Engine called Tau T2A. These VMs are the latest addition to the Tau VM family that offers VMs optimized for cost-effective performance for scale-out workloads and are available in preview.
The company added Tau VMs to the Compute Engine based on AMD’s 3rd Generation EPYC Processor (codenamed Milan) and claims up to 42% higher price-performance compared to general-purpose VMs of any of the leading public cloud vendors. The Tau T2A VM is a new addition to the Tau VM and is based on the Arm architecture. It is powered by Ampere Altra Arm-based processors and comes in multiple predefined VM shapes, with up to 48 vCPUs per VM and 4GB of memory per vCPU.
Furthermore, it offers 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.
T2A VMs support the 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. Ishan Sharma, senior product manager, Google Kubernetes Engine (GKE), wrote in a Google Cloud blog post on Tau T2A VMs and GKE:
We are also thrilled to announce that you can run your containerized workloads on the Arm architecture using GKE. Arm nodes come packed with the key GKE features you love on the x86 architecture, including the ability to run in GKE Autopilot mode for a hands-off experience or on GKE Standard clusters where you manage your own node pools.
In addition, developers building applications on Google Cloud can now use several Google Cloud services with T2A VMs besides GKE, using the recently launched Batch service to run batch jobs on T2A instances and Dataflow workloads, benefiting from cost savings and minimizing processing time. Google also intends to bring integrations with other Google services later this year.
Other public cloud providers such as Azure and AWS also offer a wide selection of instance types, varying combinations of storage, CPU, memory, and networking capacity, allowing customers to scale resources to the requirements of their target workload. Some instances run on Arm-based processors, for example, the AWS C7g EC2 instances running on the Arm-based Graviton3 processors or the Azure Dpsv5 VM-series (preview) running on Ampere Altra Arm-based processors.
In another Google Cloud blog post, Subra Chandramouli and Jamie Kinney, both product managers at Google, provided a reason for the Tau T2A VM:
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.
However, a respondent on the Hacker News thread on the preview release of Tau T2A VMs said:
I'm glad to see this, but I'm disappointed that this tops out at 48 vCPUs. The equivalent from Amazon has 64, and Altra processors support up to 128.
Currently, the TAU T2A VMs are in preview in several Google Cloud regions: US-central (Iowa - Zone A,B,F), Europe-west4 (Netherlands - Zone A,B), and Asia-southeast1 (Singapore - Zone B,C) and will be in general availability in the coming months. Pricing-wise, the VMs are price-performance optimized, for instance, a 32vCPU VM with 128GB RAM will be priced at $1.232 per hour for on-demand usage in US-central1. More details on pricing are available on the pricing page.