See this document in CiteSeerX!

Two Computer Systems Paradoxes: Serialize-to-Parallelize, and Queuing Concurrent-Writes (1995)  (Make Corrections)  (5 citations)
Rimon Orni, Uzi Vishkin



  Home/Search   Context   Related

 
View or download:
umd.edu/users/vishkin/PUBLIC...rimon.ps
Cached:  PS.gz  PS  PDF   Image  Update  Help

From:  umd.edu/users/vishkin/PU...papers (more)
Homepages:  U.Vishkin  

Rate this article: (best)
  Comment on this article  
(Enter summary)

Abstract: We present and examine the following Serialize-to-Parallelize Paradox: suppose a programmer has a parallel algorithm in mind; the programmer must serialize the algorithm, and is actually trained to suppress its parallelism, while writing code; later, however, compilation and runtime techniques are used to reverse the results of this serialization effort and extract as much parallelism as possible. This work actually provides examples where parallel or parallel-style code enables extracting more ... (Update)

Context of citations to this paper:   More

.... project entitled SUIF (Stanford University Intermediate Format) in the context of multiprocessors [SUIF] A recent position paper, [OV95], criticizes adverse effects in current programming practice which block, and even undoes, potential benefits from parallel algorithmic...

.... found in the problem from the compiler since parallelism is a resource which can lead to e ciency, even in sequential execution [13, 19]. What objectives should be set for the design of algorithms and programs Let us assume that producing correct solutions is not the...

Cited by:   More
Graph and Hashing Algorithms for Modern Architectures.. - Black, Jr., Martel, Qi (1998)   (Correct)
Designing Fast Graph Data Structures: An Experimental Approach - Jr., Martel (1998)   (Correct)
Primitives of Sequential and Parallel Computation - Forsell, Leppänen, Penttonen (1998)   (Correct)

Active bibliography (related documents):   More   All
0.5:   Thinking in Parallel: Some Basic Data-Parallel Algorithms and.. - Vishkin (2002)   (Correct)
0.1:   NESL: A Nested Data-Parallel Language (Version 3.1) - Blelloch (1995)   (Correct)
0.1:   NESL: A Nested Data-Parallel Language (Version 2.6) - Blelloch (1993)   (Correct)

Similar documents based on text:   More   All
0.1:   More Efficient Object Serialization - Philippsen, Haumacher (1999)   (Correct)
0.1:   Why Object Serialization is Inappropriate for Providing.. - Evans (2000)   (Correct)
0.1:   A More Efficient RMI for Java - Nester, Philippsen, Haumacher (1999)   (Correct)

Related documents from co-citation:   More   All
4:   The Stanford GraphBase (context) - Knuth - 1993
3:   DIMACS Series in Discrete Mathematics and Theoretical Computer Science (context) - Johnson, Trick et al.
3:   Experimental study of minimum cut algorithms - Chekuri, Goldberg et al. - 1997

BibTeX entry:   (Update)

R. Orni and U. Vishkin. Two Computer Systems Paradoxes: Serialize-to-Parallelize, and Queuing Concurrent-Writes. Preprint, September 1995. http://citeseer.ist.psu.edu/orni95two.html   More

@techreport{ orni95two,
    author = "Rimon Orni and Uzi Vishkin",
    title = "Two Computer Systems Paradoxes: Serialize-to-Parallelize, and Queuing Concurrent-Writes",
    number = "CS-TR-3586",
    year = "1995",
    url = "citeseer.ist.psu.edu/orni95two.html" }
Citations (may not include all citations):
1575   Computer Architecture A Quantitative Approach (context) - Hennessy, Patterson - 1990
1575   Computer Architecture A Quantitative Approach (context) - Hennessy, Patterson - 1994
835   High performance Fortran language specification - Fortran - 1993
480   The program dependence graph and its uses in optimization (context) - Ferrante, Ottenstein et al. - 1987
353   Software pipelining: An effective scheduling technique for v.. (context) - Lam - 1988
314   An Introduction to Parallel Algorithms (context) - J'aJ'a - 1992
230   Limits of instruction-level parallelism - Wall - 1993
157   Limits of control flow on parallelism - Lam, Wilson - 1992
130   Implementation of a portable nested data-parallel language - Blelloch, Chatterjee et al. - 1993
84   Reducing the cost of branches (context) - McFarling, Hennessy - 1986
77   parallel connectivity algorithm (context) - Shiloach, Vishkin et al. - 1982
75   Measuring parallelism in computation-intensive scientific/en.. (context) - Kumar - 1988
69   Compiling collection-oriented languages onto massively paral.. - Blelloch, Sabot - 1990
67   language for data parallel programming (context) - Rose, Steele et al. - 1987
58   Dynamic dependency analysis of ordinary programs - Austin, Sohi - 1992
53   The QRQW PRAM: Accounting for contention in parallel algorit.. (context) - Gibbons, Matias et al. - 1994
46   Tracing with pixie (context) - Smith - 1991
35   The impact of synchronization and granularity on parallel sy.. - Chen, Su et al. - 1990
31   Comparing software and hardware schemes for reducing the cos.. - Hwu, Conte et al. - 1989
27   parallel max-flow algorithm (context) - Shiloach, Vishkin et al. - 1982
25   Power and PowerPC (context) - Weiss, Smith - 1994
23   The Stanford GraphBase (context) - Knuth - 1993
16   Can parallel algorithms enhance serial implementation - Vishkin - 1994
9   Software pipelining: An evaluation of enhanced pipelining (context) - Jones, Allan - 1991
2   Parallel Processing using Depth-First Search and Breadth-Fir.. (context) - Eckstein - 1977
1   Ultra-pipelining: An efficient software pipelining algorithm (context) - Krishnamurthy - 1993



The graph only includes citing articles where the year of publication is known.


Documents on the same site (http://www.umiacs.umd.edu/users/vishkin/PUBLICATIONS/papers.html):   More
From Algorithm Parallelism to Instruction-Level Parallelism: An.. - Vishkin (1997)   (Correct)
Biconnectivity Approximations and Graph Carvings - Khuller, Vishkin (1994)   (Correct)
On Parallel Hashing and Integer Sorting - Matias, Vishkin (1991)   (Correct)

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