(Enter summary)
Abstract: Time-sharing operating systems may delay application processing
of incoming messages because other processes are scheduled when the messages
arrive. In this paper, we present a simple adjustment to application
polling behavior that reduces this effect when the parallel process competes
with CPU-intensive sequential jobs. Our results showed that moderate improvement
might be achieved. However, the effectiveness of the approach is
highly application-dependent. (Update)
Active bibliography (related documents): More All
0.0: Parallel Job Scheduling on Heterogeneous Networks of.. - Lynch
(Correct)
0.0: Coscheduling under Memory Constraints in a NOW Environment - Francesc Gin Francesc
(Correct)
0.0: A Comparative Evaluation of Implicit Coscheduling.. - Cosimo Anglano.. (2000)
(Correct)
Similar documents based on text: More All
0.4: Multi-threading and Remote Latency in Software DSMs - Thitikamol, Keleher (1997)
(Correct)
0.3: Thread Migration and Communication Minimization in DSM Systems - Thitikamol, Keleher (1999)
(Correct)
0.3: Optimizing Message Delivery in Asynchronous.. - Sharma.. (1998)
(Correct)
BibTeX entry: (Update)
@inproceedings{ thitikamol99communicationintensive,
author = "Kritchalach Thitikamol and Peter J. Keleher",
title = "Communication-Intensive Parallel Applications and Non-dedicated Clusters of Workstations",
booktitle = "{IPPS}/{SPDP} Workshops",
pages = "511-518",
year = "1999",
url = "citeseer.ist.psu.edu/492610.html" }
Citations (may not include all citations):
353
The SPLASH-2 Programs: Characterization and Methodological C..
- Woo
80
The Relative Importance of Concurrent Writers and Weak Consi..
- Keleher
80
Effective Distributed Scheduling of Parallel Workloads
- Dusseau, Arpaci et al. - 1996
40
Dynamic coscheduling on workstation clusters
- Sobalvarro - 1998
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