FlexDirect Analytics – Insight tool for your GPU cluster utilization
As organizations are quickly adopting machine learning, AI and data science workloads, they are increasingly deploying accelerated compute hardware (such as GPUs, FPGAs and AI ASICs). However, the new infrastructure is deployed with no virtualization, sharing nor any abstraction layer (all of which are heavily utilized with CPU servers). This causes AI servers’ (e.g. GPUs) utilization and efficiency to be very low. FlexDirect Analytics is a passive software sensor and monitoring tool, that allows the infrastructure manager to gain insights and visibility on the utilization of the GPU servers. FlexDirect Analytics provides time series data of utilization and efficiency of each physical GPU in your network, without any impact or change to your workloads, workflow or deployment. Utilization and efficiency can be exported for analysis with off-line tools. FlexDirect Analytics has a very low memory and compute footprint and is a lightweight process. It runs in public cloud (with GPU instances), private cloud and with any known hardware supplier of GPU servers.