The Community for Technology Leaders
2013 IEEE 19th International Symposium on High Performance Computer Architecture (HPCA) (2001)
Nuevo Leone, Mexico
Jan. 20, 2001 to Jan. 24, 2001
ISBN: 0-7695-1019-1
pp: 0097
Manuel E. Acacio , Universidad de Murcia
José González , Universidad de Murcia
José M. García , Universidad de Murcia
José Duato , Universidad Polit?nica de Valencia
Abstract: The memory overhead introduced by directories constitutes a major hurdle in the scalability of cc-NUMA archiectures, which makes the shared-memory paradigm unfeasible for very large-scale systems. This work is focused on improving the scalability of shared-memory multiprocessors by significantly reducing the size of the directory. We propose multilayer clustering as an effective approach to reduce the directory-entry width. Detailed evaluation for 64 processors shows that using this approach we can drastically reduce the memory overhead, while suffering a performance degradation very similar to previous compressed schemes (such as Coarse Vector). In addition, a novel two-level directory architecture is proposed in order to eliminate the penalty caused by these compressed directories. This organization consists of a small Full-Map first- level directory (which provides precise information for the most recently referenced lines)and a compressed second- level directory (which provides in-excess information). Results show that a system with this directory architecture can achieve the same performance as a multiprocessor with a big and non-scalable Full-Map directory, with a very significant reduction of the memory overhead.
Manuel E. Acacio, José González, José M. García, José Duato, "A New Scalable Directory Architecture for Large-Scale Multiprocessors", 2013 IEEE 19th International Symposium on High Performance Computer Architecture (HPCA), vol. 00, no. , pp. 0097, 2001, doi:10.1109/HPCA.2001.903255
358 ms
(Ver 3.3 (11022016))