Atomic Vector Operations on Chip Multiprocessors
Sanjeev Kumar, Daehyun Kim, Mikhail Smelyanskiy et al. · ACM SIGARCH Computer Architecture News · 2008
The current trend is for processors to deliver dramatic improvements in parallel performance while only modestly improving serial performance. Parallel performance is harvested through vector/SIMD instructions as well as multithreading (through both multithreaded cores and chip multiprocessors). Vector parallelism can be more efficiently supported than multithreading, but is often harder for software to exploit. In particular, code with sparse data access patterns cannot easily utilize the vector/SIMD instructions of mainstream processors. Hardware to scatter and gather sparse data has previously been proposed to enable vector execution for these codes. However, on multithreaded architectures, a number of applications spend significant time on atomic operations (e.g., parallel reductions), which cannot be vectorized using previously proposed schemes. This paper proposes architectural support for atomic vector operations (referred to as GLSC) that addresses this limitation. GLSC extends scatter-gather hardware to support atomic memory operations. Our experiments show that the GLSC provides an average performance improvement on a set of important RMS kernels of 54% for 4-wide SIMD.
Cited by 27 works. doi:10.1145/1394608.1382154
Authors
- Sanjeev Kumar
- Daehyun Kim
- Mikhail Smelyanskiy
- Yen-Kuang Chen
- Jatin Chhugani
- Christopher J. Hughes
- Changkyu Kim
- Anthony D. Nguyen
References
- Marching cubes: A high resolution 3D surface construction algorithm
- Marching cubes: A high resolution 3D surface construction algorithm
- A new approach to the maximum-flow problem
- The PARSEC benchmark suite
- Transactional memory
- Basic Techniques for the Efficient Coordination of Very Large Numbers of Cooperating Sequential Processors
- The SGI Origin
- Real-Time Collision Detection
- Synchronization and communication in the T3E multiprocessor
- Chip multiprocessing and the cell broadband engine