January 22, 2021
Sunnyvale, CA Tensor Networks’ Matrix-OS ver 8.3 includes over 60 updates and easier installation and provisioning processes for end-users and system network administrators.
One of the main features in Matrix-OS ver. 8.3 is the tie-in to CentOS streams for access to upstream patches and features.
In addition, another primary feature is the auto-recognition of acceleration technologies for high-performance computing co-processors like NVIDIA’s GPU.
With this new functionality if an NVIDIA GPU is detected during installation MatrixOS Ver. 8.3 will automatically implement drivers, settings and the development environment.
The end-user is no longer required to need to figure out how to get their accelerators working wasting project time and costs. Nor will they be required to get into the command line terminal to implement services for the accelerator.
Now that these tasks are automated the end-user can get right to their specific applications and services for their work loads. The Matrix-Data Science Router is based upon an NVIDIA EGX reference design, as well as an Intel uCPE reference design. This makes it an ideal platform to leverage NVIDIA’s NGC Cloud Service for applications like machine learning, artificial intelligence, and Data Science.
With Matrix-OS ver. 8.3 end-users save time and cost running their HPC workloads on the Matrix-Data Science Router Platform.
This is some of our automation technology at work!