Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Arm chip architecture received another boost of support from Tetrate. The company, which is building service mesh support software, announced that their products will now support Arm’s Neoverse platform. Both of Tetrate’s major tools, Envoy and Istio, now run natively on the Arm chips that are rapidly becoming known for offering cheaper computing in cloud environments.
In addition, Tetrate is rolling out a hardened version of its code to support defense clients. Some federal customers want a version of Tetrate’s Istio to support Kubernetes and other services in various military clouds like the Platform One program and Iron Bank.
Tetrate’s tools simplify the job of linking together large networks of smaller applications that work conjointly. The idea of breaking up a software application into smaller, more manageable pieces known as microservices has become the dominant architecture.
Rise in demand for microservices in the enterprise
It is now common for enterprise packages to be composed of dozens, hundreds or maybe even thousands of smaller microservices. These microservices need to coordinate and that’s where Tetrate comes in. Their software handles the chores of communication and organizing the individual services into what it calls a “mesh.”
The AI Impact Tour
Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you!
Tetrate’s software will encrypt messages as they travel between microservice providers, protecting any personal information along the way. The software will also authenticate the different nodes in the mesh, simplifying the job of ensuring security.
The Arm chips have been gaining popularity in cloud computing because they offer better performance at a lower price point. Nvidia, for instance, is making Arm-based chips with Arm cores and dubbing them “DPUs” or data processing units.
Several companies like Apple use the architecture because it can offer faster performance, while using less energy. Cloud vendors like Amazon have different needs than phone manufacturers, but saving electricity is a big focus for them.
Arm chip gains traction
In their announcement, Tetrate targeted the new Gravitron processors launched by AWS and highlighted their faster benchmark scores. The new Graviton3 is said to offer 25% better performance than the earlier Gravitron2, but some jobs may see even more. Some machine learning training algorithms, for instance, rely heavily on single precision floating-point calculations and AWS estimates that the Graviton3 may double the speed for these particular tasks.
“We are just like another workload running on these chips, so the reduction [in cost and run time] will be similar to any other workloads they see,” explained Varun Talwar, a cofounder of Tetrate. “It ranges anywhere from 10% to 35% depending on the situation,” he claims.
While Tetrate partnered with both Arm and AWS in this particular project, the Arm chip is finding traction in other deployments. Talwar says that Tetrate was motivated by seeing the Arm chips appear in some of the hardware running in edge machines close to the users.
“Those projects tend to use lighter weight Kubernetes and they are very sensitive to compute time, cost and latency.” said Talwar.
The federal customers share all of these needs with increased worries about security. The new hardened versions of Tetrate tools will also support Arm architectures.
“For us to have a secure build where they are actually using it for authentication, authorization and encryption for every service to service communication and to run that at scale is quite a significant achievement,” explained Talwar.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.