by Hubertus Franke, Jose E. Moreira, Pratap Pattnaik
http://www.cs.huji.ac.il/~feit/parsched/p-99-8.ps.gz
Add To MetaCart
Abstract:
Abstract. Job management subsystems in parallel environments have to address two important issues: (i) how to associate processes present in the system to the tasks of parallel jobs, and (ii) how to control execution of these tasks. The standard UNIX mechanism for job control, process groups, is not appropriate for this purpose as processes can escape their original groups and start new ones. We introduce the concept of genealogy, in which a process is identified by the genetic footprint it inherits from its parent. With this concept, tasks are defined by sets of processes with a common ancestor. Process tracking is the mechanism by which we implement the genealogy concept in the IBM AIX operating system. No changes to the kernel are necessary and individual process control is achieved through standard UNIX signaling methods. Performance evaluation, on both uniprocessor and multiprocessor systems, demonstrate the efficacy of job control through process tracking. Process tracking has been incorporated in a research prototype gang-scheduling system for the IBM RS/6000 SP. 1
Citations
|
799
|
Condor - A Hunter of Idle Workstations
– Litzkow, Livny, et al.
- 1988
|
|
293
|
Network Time Protocol (Version 3): Specification, Implementation and Analysis", RFC 1305
– Mills
- 1992
|
|
240
|
Scheduling Techniques for Concurrent Systems
– Ousterhout
- 1982
|
|
188
|
The NAS Parallel Benchmarks 2.0
– Bailey, Harris, et al.
- 1995
|
|
136
|
A Survey of Scheduling in Multiprogrammed Parallel Systems
– Feitelson
- 1994
|
|
110
|
The ANL/IBM SP scheduling system
– Lifka
- 1995
|
|
107
|
UNIX Internals, The New Frontiers
– Vahalia
- 1996
|
|
101
|
Effective Distributed Scheduling of Parallel Workloads
– Dusseau, Arpaci, et al.
- 1996
|
|
90
|
Improved utilization and responsiveness with gang scheduling
– Feitelson, Jette
- 1997
|
|
70
|
Demand-based Coscheduling of Parallel Jobs on Multiprogrammed Multiprocessors
– Sobalvarro
- 1997
|
|
48
|
Dynamic Coscheduling on Workstation Clusters
– Sobalvarro, Pakin, et al.
- 1998
|
|
35
|
Adaptive algorithms for managing a distributed data processing workload
– Aman, Eilert, et al.
- 1997
|
|
32
|
Evaluation of design choices for gang scheduling using distributed hierarchical control
– Feitelson, Rudolph
- 1996
|
|
16
|
The EASY-Loadleveller API Project
– Skovira, Chan, et al.
- 1996
|
|
12
|
Improving First-Come-First-Serve Job Scheduling by Gang Scheduling
– Schwiegelshohn, Yahyapour
- 1998
|
|
11
|
Time sharing massively parallel machines
– Gorda, Wolski
- 1995
|
|
11
|
Extensible Resource Management for Cluster Computing
– Islam, Prodromidis, et al.
- 1996
|
|
8
|
An infrastructure for efficient parallel job execution in terascale computing environments
– Moreira, Chan, et al.
- 1998
|
|
6
|
Implementing the Combination of Time Sharing and Space Sharing on AP/Linux
– Suzaki, Walsh
- 1998
|
|
5
|
Expanding symmetric multiprocessor capability through gang scheduling
– Jette
- 1998
|
|
1
|
Gang Scheduling for Highly Efficient Multiprocessors
– Franke, Pattnaik, et al.
- 1996
|
