Accelerating Graph and Machine Learning Workloads Using a Shared Memory Multicore Architecture with Auxiliary Support for In-hardware Explicit Messaging.
Halit DoganFarrukh HijazMasab AhmadBrian KahnePeter WilsonOmer KhanPublished in: IPDPS (2017)
Keyphrases
- shared memory
- parallel architecture
- multi processor
- memory access
- multithreading
- parallel algorithm
- message passing
- parallel computing
- parallel programming
- parallel architectures
- distributed memory
- multi core processors
- interprocess communication
- address space
- parallel computers
- parallel computation
- commodity hardware
- graphic processing unit
- real time
- parallel machines
- computer systems
- parallel execution
- heterogeneous platforms
- processing units
- hardware design
- access patterns
- fault tolerant
- image processing
- massively parallel
- computing systems
- parallel processing
- graph cuts
- program execution
- bayesian networks
- database systems
- computer vision