A scalable approach to solving dense linear algebra problems on hybrid CPU-GPU systems
- Indiana Univ.-Purdue Univ., Indianapolis, IN (United States)
- Univ. of Tennessee, Knoxville, TN (United States); Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Univ. of Manchester (United Kingdom)
Aiming to fully exploit the computing power of all CPUs and all graphics processing units (GPUs) on hybrid CPU-GPU systems to solve dense linear algebra problems, in this paper we design a class of heterogeneous tile algorithms to maximize the degree of parallelism, to minimize the communication volume, and to accommodate the heterogeneity between CPUs and GPUs. The new heterogeneous tile algorithms are executed upon our decentralized dynamic scheduling runtime system, which schedules a task graph dynamically and transfers data between compute nodes automatically. The runtime system uses a new distributed task assignment protocol to solve data dependencies between tasks without any coordination between processing units. By overlapping computation and communication through dynamic scheduling, we are able to attain scalable performance for the double-precision Cholesky factorization and QR factorization. Finally, our approach demonstrates a performance comparable to Intel MKL on shared-memory multicore systems and better performance than both vendor (e.g., Intel MKL) and open source libraries (e.g., StarPU) in the following three environments: heterogeneous clusters with GPUs, conventional clusters without GPUs, and shared-memory systems with multiple GPUs.
- Research Organization:
- Indiana Univ.-Purdue Univ., Indianapolis, IN (United States); Univ. of Tennessee, Knoxville, TN (United States); Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE
- Contributing Organization:
- Univ. of Manchester (United Kingdom)
- Grant/Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1361295
- Journal Information:
- Concurrency and Computation. Practice and Experience, Vol. 27, Issue 14; ISSN 1532-0626
- Publisher:
- WileyCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Scaling Up Parallel Computation of Tiled QR Factorizations by a Distributed Scheduling Runtime System and Analytical Modeling
|
journal | March 2018 |
Tiling-Based Programming Model for Structured Grids on GPU Clusters
|
conference | January 2020 |
Similar Records
Batched matrix computations on hardware accelerators based on GPUs
Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)