Verge.io virtualizes GPU computing to reduce complexity and cost

Verge.io, the company that virtualizes data centers, is adding new features to its Verge-OS software to bring users the performance of GPUs as virtualized and shared resources.

The move creates a cost-effective, simple, and flexible way to perform GPU-based machine learning, remote desktop, and other compute-intensive workloads in an agile, scalable Verge-OS virtual data center, according to the vendor. and secure.

Verge-OS extracts compute, network, and storage from commodity servers and creates pools of raw resources that are easy to run and manage, creating feature-rich infrastructures for environments and workloads like clustered HPC in universities, ultra-converged and hyper-converged enterprises, DevOps and Test/Dev, compliant medical and healthcare, remote and edge computing, including VDI, and xSP offering hosted services, including private clouds.

Rather than provisioning GPUs across the organization, Verge.io allows users and applications with access to a virtual data center to share computing resources from a single GPU-equipped server. Users/administrators can “port” an installed GPU to a virtual datacenter by simply creating a virtual machine that has access to that GPU and its resources.

Alternatively, Verge.io can handle GPU virtualization and serve vGPUs to virtual data centers. This allows organizations to easily manage vGPUs on the same platform as all other shared resources.

“The ability to deploy a GPU in a virtualized, converged environment and access that performance as needed, even remotely, radically reduces hardware investment while simplifying management,” said Yan Ness, CEO of Verge .io. “Our users increasingly need GPU performance, from scientific research to machine learning, so vGPU and GPU Passthrough are easy ways to share and pool GPU resources like they do with the rest of their capabilities. treatment.”

Verge-OS is ultra-lightweight (less than 300,000 lines of code) software that is easy to install and scale on low-cost, self-managed AI/ML-based commodity hardware. A single license replaces separate hypervisor, networking, storage, data protection, and management tools to simplify operations and reduce the size of complex technology stacks.

Verge-OS based secure virtual data centers include all enterprise data services such as global deduplication, disaster recovery, continuous data protection, snapshots, long distance synchronization and automatic failover.

They are ideal for creating honeypots, sandboxes, cyber ranges, isolated computes, and secure compliance enclaves to meet regulations such as HIPAA, CUI, SOX, NIST, and PCI. Nested multitenancy gives service providers, departmental enterprises, and campuses the ability to assign resources and services to groups and subgroups.

For more information on these updates, visit www.verge.io.

Maria H. Underwood