메뉴 건너뛰기




Volumn 2, Issue 1, 2004, Pages 22-35

Dynamic tiling for effective use of shared caches on multithreaded processors

Author keywords

compilers; memory hierarchies; multithreaded processors; operating systems; runtime systems

Indexed keywords


EID: 35448980346     PISSN: 17400562     EISSN: 17400570     Source Type: Journal    
DOI: 10.1504/IJHPCN.2004.009265     Document Type: Article
Times cited : (7)

References (29)
  • 5
    • 0031223129 scopus 로고    scopus 로고
    • Compiler blockability of dense matrix factorisations
    • Carr, S. and Lehoucq, R. (1997) ‘Compiler blockability of dense matrix factorisations’, ACM Transactions on Mathematical Software, Vol. 23, No. 3, pp.336–361.
    • (1997) ACM Transactions on Mathematical Software , vol.23 , Issue.3 , pp. 336-361
    • Carr, S.1    Lehoucq, R.2
  • 7
    • 77953936946 scopus 로고    scopus 로고
    • Intel hyper-threading technology
    • Cross, R. (2002) ‘Intel hyper-threading technology’, Intel Technology Journal, Vol. 6, No. 1.
    • (2002) Intel Technology Journal , vol.6 , Issue.1
    • Cross, R.1
  • 8
    • 0037230301 scopus 로고    scopus 로고
    • High performance linear algebra algorithms using new generalized data structures and matrices
    • Gustavson, F. (2002) ‘High performance linear algebra algorithms using new generalized data structures and matrices’, IBM Journal of Research and Development, Vol. 17, No. 1, pp.31–56.
    • (2002) IBM Journal of Research and Development , vol.17 , Issue.1 , pp. 31-56
    • Gustavson, F.1
  • 10
    • 0031199614 scopus 로고    scopus 로고
    • Converting thread-level parallelism to instruction-level parallelism via simultaneous multithreading
    • Lo, J., Emer, J., Levy, H., Stamm, R., Tullsen, D. and Eggers, S. (1997) ‘Converting thread-level parallelism to instruction-level parallelism via simultaneous multithreading’, ACM Transactions on Computer Systems, Vol. 15, No. 3, pp.322–353.
    • (1997) ACM Transactions on Computer Systems , vol.15 , Issue.3 , pp. 322-353
    • Lo, J.1    Emer, J.2    Levy, H.3    Stamm, R.4    Tullsen, D.5    Eggers, S.6
  • 27
    • 84951724411 scopus 로고    scopus 로고
    • Currently, Linux assigns different processor IDs to the hardware threads on an SMT processor, as if each hardware thread were a CPU on its own
    • Currently, Linux assigns different processor IDs to the hardware threads on an SMT processor, as if each hardware thread were a CPU on its own.
  • 28
    • 84951724412 scopus 로고    scopus 로고
    • In Linux, each kernel thread has a unique identifier
    • In Linux, each kernel thread has a unique identifier.
  • 29
    • 84951724413 scopus 로고    scopus 로고
    • All reported averages are arithmetic means taken from the executions of a program with a fixed number of threads across all dataset sizes
    • All reported averages are arithmetic means taken from the executions of a program with a fixed number of threads across all dataset sizes.


* 이 정보는 Elsevier사의 SCOPUS DB에서 KISTI가 분석하여 추출한 것입니다.