Building Heterogeneous Unified Virtual Memories (UVMs) without the Overhead
2016 (English)In: ACM Transactions on Architecture and Code Optimization (TACO), ISSN 1544-3566, E-ISSN 1544-3973, Vol. 13, no 1, 1Article in journal (Refereed) Published
This work proposes a novel scheme to facilitate heterogeneous systems with unified virtual memory. Research proposals implement coherence protocols for sequential consistency (SC) between central processing unit (CPU) cores and between devices. Such mechanisms introduce severe bottlenecks in the system; therefore, we adopt the heterogeneous-race-free (HRF) memory model. The use of HRF simplifies the coherency protocol and the graphics processing unit (GPU) memory management unit (MMU). Our protocol optimizes CPU and GPU demands separately, with the GPU part being simpler while the CPU is more elaborate and latency aware. We achieve an average 45% speedup and 45% energy-delay product reduction (20% energy) over the corresponding SC implementation.
Place, publisher, year, edition, pages
2016. Vol. 13, no 1, 1
Multicore; heterogeneous coherence; GPU MMU design; virtual coherence protocol; directory-less protocol
IdentifiersURN: urn:nbn:se:uu:diva-295765DOI: 10.1145/2889488ISI: 000373904600001OAI: oai:DiVA.org:uu-295765DiVA: diva2:934693
FunderEU, FP7, Seventh Framework Programme, FP7-ICT-288653EU, European Research Council, TIN2012-38341-C04-03