Y. Zhu and L. J. Hendren. Locality Analysis for Parallel C Programs. IEEE Transactions on Parallel and Distributed Systems, 10(2):99--114, February 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....significantly more flexible sharing of data is possible, allowing for more data to be statically identified as private and thereby making privacy dependent analyses more effective. EARTH C explicitly offers both local global and shared private type qualifiers. Local global may be inferred [36], but shared private must be given explicitly [18] Our approach shows that shared private is amenable to inference as well, operating either fully automatically or to augment programmer directives. The broader EARTH C project has also clearly demonstrated the value of identifying local private ....

Y. Zhu and L. Hendren. Locality analysis for parallel C programs. IEEE Transactions on Parallel and Distributed Systems, 10(2):99--114, Feb. 1999.

....access latencies, data locality is potentially the most important performance issue of NUMA architectures among factors like memory contention and false sharing [13] which affect performance. An improvement of data locality can be achieved through a static data distribution during compiling time [8, 10, 18] and a dynamic redistribution in run time [12, 15] In the compiler approach, locality is analyzed during the compiling procedure using the information about the context of data structures and functions. Communication among processors is minimized via initially effective data distribution[10] ....

....approach, locality is analyzed during the compiling procedure using the information about the context of data structures and functions. Communication among processors is minimized via initially effective data distribution[10] communication overlap [8] and elimination of pseudoremote accesses [18]. In the dynamic redistribution approach, the communication behavior of running programs is analyzed and virtual memory pages are dynamically migrated to the nodes that reference them more frequently. The location of a page is decided usually based on the first iteration, trace of cache misses, ....

Yingchun Zhu and Laurie J. Hendren. Locality analysis for parallel C programs. IEEE Transactions on Parallel and Distributed Systems, 10(2), February 1999.

....or on a remote memory. Remote memory accesses can be orders of magnitude slower than local memory accesses, and so it is important to reduce remote accesses as much as possible. Locality Analysis was designed to automatically determine when memory references via pointers were to local memory[ZH99a] whereas communication analysis was designed to reduce the number of remote memory references by eliminating redundant references and blocking related 4 Research Portfolio (External) references together[ZH99b] Other advances within McCAT: There are many other important facets of the compiler ....

Y. Zhu and L. Hendren. Locality analysis for parallel C programs. IEEE Transactions on Parallel and Distributed Systems, 10(2):99-114, Feb. 1999. 12 Research Portfolio (External)

....a valid EARTH C program, and the compiler automatically produces a correct low level threaded program. Usually the programmer uses the EARTH C constructs to expose coarse grain parallelism, and to add some information about data locality. The compiler performs analysis to infer additional locality[22], to expose fine grain parallelism via data dependence analysis, and to reduce communication (the topic of this paper) Two sample list processing functions, written in EARTH C, are given in Figure 1. In both cases the functions take a pointer to a list head, and a pointer to a node x, and return ....

....1(b) the call to equal node(head,x) is specified to occur at the OWNER OF(x) This means that within the body of equal node, the second parameter can be assumed to be a local pointer. This locality can be exposed either through explicit local declarations, or automatically via locality analysis [22]. 2.2 Memory Model An important point about the EARTH C language is that parallel computations expressed via parallel statement sequences or forall loops, may not interfere except on explicit shared variables. It is the programmer s responsibility to ensure this non interference for explicit ....

[Article contains additional citation context not shown here]

Yingchun Zhu and Laurie Hendren. Locality analysis for parallel C programs. In Proc. of the PACT'97, pages 2--13, San Francisco, Nov. 1997. North-Holland Pub. Co.

No context found.

Y. Zhu and L. J. Hendren. Locality Analysis for Parallel C Programs. IEEE Transactions on Parallel and Distributed Systems, 10(2):99--114, February 1999.

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