Abstract:
Load balancing allows to increase system throughput and to reduce application response time. However, when the computing system increases in scale, load balancing becomes very complex to manage and is time consuming. Moreover, actual load balancing mechanisms are not able to manage heterogeneity both in system architectures and application behaviors. To deal with such systems we propose to split load balancing between environment oriented and application oriented managers. Negotiation between managers allows to manage heterogeneous and large systems, while internal management allows to use the most suitable policy. Then, hybrid applications, using simultaneously different programming models can be executed efficiently.
Citations
|
829
|
Condor : A hunter of idle workstations
– Litzkow, Livny, et al.
- 1988
|
|
242
|
Transparent process migration: Design alternatives and the Sprite implementation
– Douglis, Ousterhout
- 1991
|
|
129
|
A Users Guide to PVM Parallel Virtual Machine
– Beguelin, Dongarra, et al.
- 1991
|
|
129
|
Lottery and Stride Schedul-ing: Flexible Proportional-Share Resource Management
– Waldspurger
- 1995
|
|
110
|
Designing a Process Migration Facility: The Charlotte Experience
– Artsy, Finkel
- 1989
|
|
67
|
The Design of a Standard Message Passing Interface for Distributed Memory
– Walker
- 1994
|
|
47
|
Multiprogramming on Multiprocessors
– Crovella, Das, et al.
- 1991
|
|
47
|
Processor allocation policies for message-passing parallel computers
– McCann, Zahorjan
|
|
39
|
Semi-Distributed Load Balancing for Massively Parallel Multicomputer Systems
– Ahmad, Ghafoor
- 1991
|
|
31
|
Utopia: a Load Sharing Facility for Large
– Zhou, Zheng, et al.
- 1993
|
|
25
|
Lsbatch: a Distributed Load Sharing Batch System
– Wang, Zhou, et al.
- 1993
|
|
23
|
Stardust: An environment for parallel programming on networks of heterogeneous workstations
– Cabillic, Puaut
- 1997
|
|
23
|
Computational field model: Toward a new computing model/methodology for open distributed environment
– Tokoro
- 1990
|
|
20
|
Nearest-neighbor algorithms for load balancing in parallel computers, Concurrency: Practice and Experience 7
– Xu, Lau, et al.
- 1995
|
|
14
|
Takeshi Horie, Satoshi Inano, Toshiyuki Shimizu, Sadayuki Kato, and Morio Ikesaka. Third generation message passing computer AP1000
– Ishihata
- 1991
|
|
10
|
Dantzig. Condor flocking: load sharing between pools of workstations
– Evers, Jongh, et al.
- 1993
|
|
10
|
AP/Linux - A modern OS for the AP1000
– Tridgell, Mackerras, et al.
- 1996
|
|
9
|
Effective load balancing in a distributed object-support operating system
– Dickman
- 1991
|
|
9
|
and Wunbutt, Load balancing with network partitioning using host groups, Parallel computing
– Evans
- 1994
|
|
7
|
Hybrid methods on network of heterogeneous parallel computers
– Edjlali, Emad, et al.
- 1994
|
|
5
|
Scalable load-sharing for distributed systems
– Kremien, Kramer, et al.
- 1993
|
|
3
|
Dealing with Heterogeneity in Stardust: An Environment for Parallel Programming on Networks of Heterogeneous Workstations
– PUAUT
- 1996
|
|
3
|
Paros: a parallel operating system for distributed memory parallel computers
– Muntean, Castro
- 1993
|
|
2
|
An interactive vizualisation environment for financial modeling on heterogeneous computing systems
– Cheng, Mills, et al.
- 1993
|
|
1
|
Ap3000 introduction. Communication
– Shiraishi
|
|
