Today, at Google NEXT’23, Google revealed the upcoming private preview of the new compute-optimized instance, C3A — powered by AmpereOne™ processors. These new C3A instances expand Google Cloud’s Ampere offerings that provide a modern, energy-efficient alternative to traditional x86 processors for their end users. This is the first AmpereOne™-based public cloud instance.
This next generation of C3A instances follows the initial collaboration with Ampere on Tau T2A instances in July’22, Google Cloud’s first compute-engine VM to run on Ampere processors. Ampere’s roadmap for rapid innovation has enabled the Google team to announce C3A instances just over a year after these initial T2A deployments. The new C3A instances will enable more broad scale adoption.
Google Cloud’s leading compute-optimized instances provide best-in-class compute, networking and storage. The C3A instances enable VMs on Ampere’s latest AmpereOne™product, delivering high core counts with single-threaded process execution, which allows for developers to scale their applications, and to do so in a predictable manner.
Ideal Cloud Native workloads for C3A instances include scale-out workloads, containerized microservices, web serving, high-performance databases, media transcoding, large-scale Java applications, cloud gaming, High-Performance Computing (HPC) and CPU based AI/ML inferencing. Ampere optimized solutions for Google Cloud are detailed in Cloud Briefs on Cassandra, Hadoop, Redis, Spark, Nginx and Media Transcoding. These Cloud Native workloads, along with runtime language support such as Java, Python, PHP, Node.js, Co, and C/C++, are ideal for Arm64 migration to Ampere processors.
Customers should expect to see Price/Performance benefits when running these Cloud Native workloads on Ampere’s Cloud Native Processor architecture. Our partners are excited to share in the journey with Ampere and Google. Here is what they have to say:
"Lightly.ai’s customers can achieve over 3x cost reduction running on Ampere-based Google Tau T2A instances on Google Cloud using Ampere AI software solutions for AI Inference. Ampere instances also provide the best-in-class solution for data preparation and data curation. Our customers have optimized costs and used the savings to invest further in development of their AI applications. At Lightly.ai we are always looking for ways to increase performance and lower costs for our customers, the next generation AmpereOne™ C3A instances on Google Cloud will deliver on this continued value proposition."
— Igor Susmelj, Co-Founder
"At Rescale, we take pride in helping our customers deliver the fastest innovation by empowering analysts, engineers, and scientists with high performance computing built for the cloud. We are pleased to power the largest full-stack platform portfolio of unique specialized architectures in the world, and our customers have already run over 100 million core hours on Arm-based architectures, including the Google Tau T2A powered by Ampere VMs. We're also excited to broaden support and range of options for this next generation of AmpereOne™ based C3A instances, which specifically provide a low energy footprint and a compelling price-performance benefit to data processing, engineering, and scientific workloads. With Ampere-based VMs on Google Cloud, we are able to offer our customers a new way to optimize spend and increase performance, while minimizing impact on the environment.”
—Joris Poort, CEO, Rescale.
"Processor innovation is happening faster than ever, and we proved that the developer experience of using new Ampere Arm based instances was easy. Not only was the experience seamless, the T2A instances delivered the highest performance we’ve seen for transactional caching performance. We expect to continue to expand the service offerings and performance benefits of these instances through our partnerships with Ampere and Google Cloud on the next generation AmpereOne™ based C3A instances.”
— Khawaja Shams, Co-Founder and CEO See Case Study >
We look forward to the journey ahead and in working closely with Google in anticipation of C3A instance general availability.
Ampere is located at booth 1632 during Google NEXT. Please visit us there to see the latest Ampere AI demos.
For more information and resources: