NVIDIA vComputeServer software is now available with Dell EMC PowerEdge servers, provides GPU virtualization to speed analytics, AI and high-performance computing.
There are 2.5 quintillion bytes of data created each day including 16 million text messages! In an effort to serve up that data faster, data center teams are increasingly turning to servers that incorporate GPUs. These accelerators, which complement the capabilities of the CPUs, use parallel processing to churn through large volumes of data at blazingly fast speeds.
But there’s a catch here. GPU-accelerated workloads, such as AI training and inferencing, often run on bare metal and may need to be managed separately, limiting utilization and flexibility. And while server virtualization is now standard in the data center, it has not necessarily included the server GPUs — until now.
Today, NVIDIA® virtual GPU (vGPU) technology is now available for the most compute-intensive data center workloads, including artificial intelligence, deep learning and high-performance computing. This leap forward is made possible by the launch of the new NVIDIA vComputeServer (vCS) software, which is making its debut at VMworld. Today, you can get vComputeServer from Dell EMC with PowerEdge servers. And if you’re not sure which PowerEdge server best fits your accelerated workload, there’s a new eBook available for you to explore the options.
vComputeServer software gives your data center team the option to run GPU workloads in virtualized environments for improved security, utilization and manageability. With vComputeServer, your team can better streamline management of GPU servers while retaining your existing workflows and potentially lowering your overall operational costs.
The software is supported on major hypervisor virtualization platforms including VMware. This means your IT team can now use the same management tools for your GPU clusters as you use for the rest of your data center.
What’s in vComputeServer?
NVIDIA vComputeServer provides features like GPU sharing, so multiple virtual machines can be powered by a single GPU, and GPU aggregation, so multiple GPUs can be allocated to a single virtual machine. This results in increased utilization and affordability.
Features of vComputeServer include:
- Advanced compute features — Error Correcting Code and Dynamic Page Retirement help prevent data corruption for high-accuracy workloads.
- Live migration — GPU-enabled virtual machines can be migrated with minimal disruption or downtime.
- Increased security —You can extend the security benefits of server virtualization to GPU clusters.
- Multi-tenant — You can isolate workloads to securely support multiple users on a single infrastructure.
- Management and monitoring — Your admin team can use the same hypervisor virtualization tools to manage GPU servers.
- License per GPU — vComputeServer is an annual subscription license purchased per GPU available with Dell EMC PowerEdge servers.
- Supported GPUs — vComputeServer is supported on NVIDIA V100 or T4 GPUs, as well as Quadro RTX 6000 and RTX 8000, and prior generations of Pascal GPUs including P40 and P100.
NVIDIA NGC Adds Support for VMware vSphere
To make this story better, NVIDIA NGC serves as a hub for GPU-optimized software for deep learning, machine learning and HPC. It offers more than 150 containers, pre-trained models, training scripts and workflows to accelerate AI from concept to production. All of these can be deployed on virtualized environments like VMware vSphere with vComputeServer.
Check out Dell EMC NGC Ready Systems including the new DSS 8440, the PowerEdge C4140 and R740 servers. These systems have been jointly validated for functionality and deliver optimized performance of AI and machine learning workloads so data scientists and developers can quickly build or scale solutions. Dell EMC’s ProSupport Plus provides single-point-of-contact support for your users and admins with direct access to NVIDIA’s experts for NGC software. That kind of support can help your team reduce risk and improve productivity.
Put it all together and you see that your IT team can now use VMware to manage more of your data center applications, including your AI applications running on systems with NVIDIA GPU accelerators.
To learn more