Without an efficient way to squeeze additional computing power from existing infrastructure, organizations are often forced to purchase additional hardware or delay projects. This can lead to longer wait times for results and potentially losing out to competitors. This problem is compounded by the rise of AI workloads which require a high GPU compute load.
ClearML has come up with what it thinks is the perfect solution to this problem – fractional GPU capability for open source users, making it possible to “split” a single GPU so it can run multiple AI tasks simultaneously.
This move recalls the early days of computing when mainframes could be shared among individuals and organizations, giving them the ability to utilize computing power without needing to buy additional hardware.
Fractional capabilities for Nvidia GPUs
ClearML says the new feature allows DevOps professionals and AI Infrastructure leaders to partition their Nvidia GTX, RTX, and data center-grade, MIG-enabled GPUs into smaller units to support multiple AI and HPC workloads, enabling users to switch between small R&D jobs and larger, more demanding training jobs.
The approach supports multi-tenancy, offering secure and confidential computing with hard memory limitation. ClearML says stakeholders can run isolated parallel workloads on a single shared compute resource, increasing efficiency and reducing costs.
“With our new free offering now supporting fractional capabilities for the broadest range of Nvidia GPUs than any other company, ClearML is democratizing access to compute as part of our commitment to help our community build better AI at any scale, faster,” says Moses Guttmann, CEO and Co-founder of ClearML. “We hope that organizations that might have a mixture of infrastructure are able to use ClearML and get more out of the compute and resources they already have.”
The new open source fractional GPU functionality is available for free on ClearML’s GitHub page.