Modeling communication in cache-coherent SMP systems

Citations Over TimeTop 10% of 2013 papers

Abstract

Most multi-core and some many-core processors implement cache coherency protocols that heavily complicate the design of optimal parallel algorithms. Communication is performed implicitly by cache line transfers between cores, complicating the understanding of performance properties. We developed an intuitive performance model for cache-coherent architectures and demonstrate its use with the currently most scalable cache-coherent many-core architecture, Intel Xeon Phi. Using our model, we develop several optimal and optimized algorithms for complex parallel data exchanges. All algorithms that were developed with the model beat the performance of the highly-tuned vendor-specific Intel OpenMP and MPI libraries by up to a factor of 4.3. The model can be simplified to satisfy the tradeoff between complexity of algorithm design and accuracy. We expect that our model can serve as a vehicle for advanced algorithm design.

Related Papers

• → Location cache(2004)28 cited
• Snoopy and Directory Based Cache Coherence Protocols: A Critical Analysis(2010)
• → Hybrid-way Cache for Mobile Processors(2011)3 cited
• → Composite Pseudo-Associative Cache for Mobile Processors(2010)2 cited
• → Composite Pseudo Associative Cache with Victim Cache for Mobile Processors(2011)