A Hybrid Task Graph Scheduler for High Performance Image Processing Workflows

作者：Blattner Timothy^*; Keyrouz Walid; Bhattacharyya Shuvra S; Halem Milton; Brady Mary

来源：Journal of Signal Processing Systems for Signal Image and Video Technology, 2017, 89(3): 457-467.

DOI：10.1007/s11265-017-1262-6

• Heterogeneous architectures
• Task graph
• Hybrid workflows
• Dataflow
• Image processing
• Matrix multiplication

Designing applications for scalability is key to improving their performance in hybrid and cluster computing. Scheduling code to utilize parallelism is difficult, particularly when dealing with data dependencies, memory management, data motion, and processor occupancy. The Hybrid Task Graph Scheduler (HTGS) improves programmer productivity when implementing hybrid workflows for multi-core and multi-GPU systems. The Hybrid Task Graph Scheduler (HTGS) is an abstract execution model, framework, and API that increases programmer productivity when implementing hybrid workflows for such systems. HTGS manages dependencies between tasks, represents CPU and GPU memories independently, overlaps computations with disk I/O and memory transfers, keeps multiple GPUs occupied, and uses all available compute resources. Through these abstractions, data motion and memory are explicit; this makes data locality decisions more accessible. To demonstrate the HTGS application program interface (API), we present implementations of two example algorithms: (1) a matrix multiplication that shows how easily task graphs can be used; and (2) a hybrid implementation of microscopy image stitching that reduces code size by ae 43% compared to a manually coded hybrid workflow implementation and showcases the minimal overhead of task graphs in HTGS. Both of the HTGS-based implementations show good performance. In image stitching the HTGS implementation achieves similar performance to the hybrid workflow implementation. Matrix multiplication with HTGS achieves 1.3x and 1.8x speedup over the multi-threaded OpenBLAS library for 16k x 16k and 32k x 32k size matrices, respectively.

• 出版日期2017-12