| P. Keleher. A High Level Abstraction of Shared Accesses. ACM Transactions on Computer Systems, Vol. 18, No. 1, pp. 1--36. Feburary 2000. |
....placement for each phase in the program, using only implicit memory reference information available from the hardware counters. UPMlib uses a mechanism called record replay to address the aforementioned problem. This mechanism is conceptually similar to the record replay barriers described in [6]. The record replay mechanism handles effectively strictly iterative parallel codes in which the same memory reference trace is repeated for a number of iterations. For noniterative programs, or iterative programs with non repetitive access patterns, UPMlib employs the sampling mechanism outlined ....
P. Keleher. A High Level Abstraction of Shared Accesses. ACM Transactions on Computer Systems, Vol. 18, No. 1, pp. 1--36. Feburary 2000.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC