Matches in SemOpenAlex for { <https://semopenalex.org/work/W1565283590> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W1565283590 abstract "Data parallelism is well-suited from algorithmic, architectural, and linguistic considerations to serve as a basis for portable parallel programming. However, the fine-grained parallelism characteristic of data-parallel programs makes the efficient implementation of such languages on MIMD machines a challenging task due to the high overheads these machines incur at small grain sizes. We claim that compile-time analysis can be used to reduce these overheads, thereby allowing data-parallel code to run efficiently on MIMD machines. This dissertation reports on the design, implementation, and evaluation of an optimizing compiler for an applicative nested data-parallel language called V scCODE. The target machine is the Encore Multimax, a coherent-cache shared-memory multiprocessor. The source language allows nested aggregate data types, and provides a variety of aggregate operations including elementwise forms, scans, reductions, and permutations. Such features greatly expand the range of applications that can be cast into a data-parallel model.We present a small set of powerful compile-time techniques that reduce the overheads on MIMD machines in several ways: by increasing the grain size of the output program, by reducing synchronization and storage requirements, and by improving locality of reference. The two key ideas behind these optimizations are the symbolic analysis of loop structures, and the hierarchical clustering of the program graph, first by loop structure, and then by loop traversal patterns. This localizes synchronization and work distribution actions to well-defined points in the output code. Loop traversal patterns are then used to identify parallel loops and to eliminate unnecessary intermediate storage. The most significant aspect of the analysis techniques is that they are symbolic in nature and work in the presence of control constructs such as conditionals and recursion.A compiler has been implemented based on these ideas and has been used to compile a large number of benchmarks. The benchmark suite covers a variety of problem domains, including dense and sparse matrix operations, tree and graph algorithms, dynamic and data-dependent algorithms, and low- and medium-level image processing applications. The performance results substantiate the claim that data-parallel code can be made to run efficiently on MIMD machines, and demonstrate that compiler optimizations are essential for good performance." @default.
- W1565283590 created "2016-06-24" @default.
- W1565283590 creator A5077666857 @default.
- W1565283590 date "1992-01-01" @default.
- W1565283590 modified "2023-09-24" @default.
- W1565283590 title "Compiling data-parallel programs for efficient execution on shared-memory multiprocessors" @default.
- W1565283590 hasPublicationYear "1992" @default.
- W1565283590 type Work @default.
- W1565283590 sameAs 1565283590 @default.
- W1565283590 citedByCount "15" @default.
- W1565283590 crossrefType "journal-article" @default.
- W1565283590 hasAuthorship W1565283590A5077666857 @default.
- W1565283590 hasConcept C11799548 @default.
- W1565283590 hasConcept C127162648 @default.
- W1565283590 hasConcept C1306188 @default.
- W1565283590 hasConcept C133875982 @default.
- W1565283590 hasConcept C140745168 @default.
- W1565283590 hasConcept C169590947 @default.
- W1565283590 hasConcept C173608175 @default.
- W1565283590 hasConcept C199360897 @default.
- W1565283590 hasConcept C21032095 @default.
- W1565283590 hasConcept C2778562939 @default.
- W1565283590 hasConcept C31258907 @default.
- W1565283590 hasConcept C41008148 @default.
- W1565283590 hasConcept C91481028 @default.
- W1565283590 hasConceptScore W1565283590C11799548 @default.
- W1565283590 hasConceptScore W1565283590C127162648 @default.
- W1565283590 hasConceptScore W1565283590C1306188 @default.
- W1565283590 hasConceptScore W1565283590C133875982 @default.
- W1565283590 hasConceptScore W1565283590C140745168 @default.
- W1565283590 hasConceptScore W1565283590C169590947 @default.
- W1565283590 hasConceptScore W1565283590C173608175 @default.
- W1565283590 hasConceptScore W1565283590C199360897 @default.
- W1565283590 hasConceptScore W1565283590C21032095 @default.
- W1565283590 hasConceptScore W1565283590C2778562939 @default.
- W1565283590 hasConceptScore W1565283590C31258907 @default.
- W1565283590 hasConceptScore W1565283590C41008148 @default.
- W1565283590 hasConceptScore W1565283590C91481028 @default.
- W1565283590 hasLocation W15652835901 @default.
- W1565283590 hasOpenAccess W1565283590 @default.
- W1565283590 hasPrimaryLocation W15652835901 @default.
- W1565283590 hasRelatedWork W1543111008 @default.
- W1565283590 hasRelatedWork W1543308214 @default.
- W1565283590 hasRelatedWork W1551170326 @default.
- W1565283590 hasRelatedWork W1583210003 @default.
- W1565283590 hasRelatedWork W1585022015 @default.
- W1565283590 hasRelatedWork W1585627840 @default.
- W1565283590 hasRelatedWork W1965122136 @default.
- W1565283590 hasRelatedWork W1996541179 @default.
- W1565283590 hasRelatedWork W2020580827 @default.
- W1565283590 hasRelatedWork W2022747918 @default.
- W1565283590 hasRelatedWork W2033214895 @default.
- W1565283590 hasRelatedWork W2087977509 @default.
- W1565283590 hasRelatedWork W2090587238 @default.
- W1565283590 hasRelatedWork W2103890844 @default.
- W1565283590 hasRelatedWork W2112708371 @default.
- W1565283590 hasRelatedWork W2156440470 @default.
- W1565283590 hasRelatedWork W2232049412 @default.
- W1565283590 hasRelatedWork W2554354717 @default.
- W1565283590 hasRelatedWork W2799001871 @default.
- W1565283590 hasRelatedWork W2979396597 @default.
- W1565283590 isParatext "false" @default.
- W1565283590 isRetracted "false" @default.
- W1565283590 magId "1565283590" @default.
- W1565283590 workType "article" @default.