Bitfusion FlexDirect on Cisco’s HyperFlex offers a seamless way for any VM to access any arbitrary fraction of GPU or multiple GPUs across the organization, anytime.

We are excited to partner with Cisco and support Cisco HyperFlex and UCS systems with FlexDirect to enable network attached GPUs on HyperFlex. Check us out at Cisco Live 2018 where we will be showcasing this solution. Now with Bitfusion, any UCS or HyperFlex installed base can be upgraded to Elastic GPU infrastructure, by simply attaching GPUs over Ethernet.

Just like CPU compute, storage and networking, GPUs can also now be disaggregated and virtualized with Bitfusion FlexDirect on Cisco HyperFlex. GPU accelerators can be partitioned into multiple virtual GPUs of any size that can be accessed remotely by VMs, over the network. With FlexDirect, GPU accelerators can be now be part of a common infrastructure resource pool and be available for use by any VM in the Cisco HyperFlex environment to enable efficient resource sharing of GPUs across organizations and applications. The same economics of the HX Data Platform now can be extended to GPUs with Bitfusion FlexDirect on HyperFlex.

Figure 1: Cisco HyperFlex Systems and Bitfusion FlexDirect

Figure 1: Cisco HyperFlex Systems and Bitfusion FlexDirect

With virtualized GPUs as part of Cisco’s HyperFlex based cloud, organizations will be able to seamlessly scale the operations with policies and business logic (time of day policies, class of users, permission to access the top performance GPUs per user class, etc.) for AI developers. GPUs from different departments can be pooled to create bigger clusters to increase compute performance and infrastructure utilization. IT gets the ability to assign GPU resources based on organization business priorities, and remotely pool together resources, while attaching them in real-time to the workloads, with known schedule and utilization plan. For example, GPU resources from Department A which completed intensive training and development schedule, can be reassigned to Department B which now experiences peak demand for GPUs for an urgent AI project.

With FlexDirect on HyperFlex, IT organizations now do not need to make hard choices between public cloud, private cloud, or on-premises; the combined solution (FlexDirect and HyperFlex) will work in any environment.

Check out our demo in the Cisco Campus in the World of Solution, in the “Unlock the Power of Data” area: Accelerate Analytics and AI [DEMO-1013] at Cisco Live 2018. You can find it on the map in the Cisco Live Mobile app using the name or demo number above.

FlexDirect operates via an intuitive command line interface (CLI) and can be integrated easily into any existing environments.

Go to to get started with FlexDirect right-away.

Robert Noakes