We are pleased to announce a new set of compelling benchmarks for Bitfusion’s production-grade Elastic AI Platform achieved in conjunction with SkyScale’s Bitfusion. It has ran several ML benchmarks with SkyScale Cloud Solutions, demonstrating that their Elastic AI via remote attached GPUs is a mainstream production-grade solution.
The below chart shows a virtual cluster as hosted on SkyScale’s cloud:
The Bitfusion GPU attached network follows a typical network disaggregation evolution… separates resources from the compute, and offers elasticity, efficiency and scale. This trend as an unavoidable evolution of heterogeneous compute, where accelerators (e.g. GPU and FPGAs) will form clusters connected to the network, and any user or workload can attach to it on-demand.
Essentially this is the epitome of Hyperconvergence or consumption-based AI. Many industry experts buy into this concept, but have asked about what the network latency impact will be. As the tests demonstrate, the impact with Bitfusion is minimal for the majority of AI use cases. Bitfusion ran several Tensorflow benchmarks, and compared local GPU (native) performance to remote performance, creating an apples to apples comparison. Here are the results:
To sum it up: Bitfusion remote on SkyScale achieves similar to native performance, with a negligible performance gap.
Free trials are available to test these benchmarks or use your own.
Click the button below, provide your contact information and we’ll get you started.