Document Type

Conference Paper

Publication Date

2016

Embargo Period

11-4-2016

Abstract

Dynamic task graph schedulers automatically balance work across processor cores by scheduling tasks among available threads while preserving dependences. In this paper, we design NabbitC, a provably efficient dynamic task graph scheduler that accounts for data locality on NUMA systems. NabbitC allows users to assign a color to each task representing the location (e.g., a processor core) that has the most efficient access to data needed during that node’s execution. NabbitC then automatically adjusts the scheduling so as to preferentially execute each node at the location that matches its color—leading to better locality because the node is likely to make local rather than remote accesses. At the same time, NabbitC tries to optimize load balance and not add too much overhead compared to the vanilla Nabbit scheduler that does not consider locality. We provide a theoretical analysis that shows that NabbitC does not asymptotically impact the scalability of Nabbit . We evaluated the performance of NabbitC on a suite of memory intensive benchmarks. Our experiments indicates that adding locality awareness has a considerable performance advantage compared to the vanilla Nabbit scheduler. In addition, we also compared NabbitC to OpenMP programs for both regular and irregular applications. For regular applications, OpenMP achieves perfect locality and perfect load balance statically. For these benchmarks, NabbitC has a small performance penalty compared to OpenMP due to its dynamic scheduling strategy. For irregular applications, where OpenMP can not achieve locality and load balance simultaneously, we find that NabbitC performs better. Therefore, NabbitC combines the benefits of locality- aware scheduling for regular applications (the forte of static schedulers such as those in OpenMP) and dynamically adapting to load imbalance (the forte of dynamic schedulers such as Cilk Plus, TBB, and Nabbit).

Share

COinS