TENSILE: A Tensor granularity dynamic GPU memory scheduling method toward multiple dynamic workloads system
This work addresses GPU memory management for systems with multiple dynamic workloads, offering an incremental improvement over existing methods.
The paper tackled the problem of GPU memory scarcity in systems with multiple dynamic workloads, such as in-database machine learning, by proposing TENSILE, a tensor-granularity dynamic GPU memory scheduling method that reduces memory peak usage. The results show that TENSILE saves more GPU memory with less overhead than prior works in both single and multiple dynamic workload scenarios.
Recently, deep learning has been an area of intense research. However, as a kind of computing-intensive task, deep learning highly relies on the scale of GPU memory, which is usually prohibitive and scarce. Although some extensive works have been proposed for dynamic GPU memory management, they are hard to apply to systems with multiple dynamic workloads, such as in-database machine learning systems. In this paper, we demonstrated TENSILE, a method of managing GPU memory in tensor granularity to reduce the GPU memory peak, considering the multiple dynamic workloads. TENSILE tackled the cold-starting and across-iteration scheduling problem existing in previous works. We implemented TENSILE on a deep learning framework built by ourselves and evaluated its performance. The experiment results show that TENSILE can save more GPU memory with less extra overhead than prior works in single and multiple dynamic workloads scenarios.