Scheduling Proxy: Enabling Adaptive-Grained Scheduling for Global Computing System

Jaesun HAN; Daeyeon PARK

doi:10.1093/ietcom/e88-b.4.1448

Scheduling Proxy: Enabling Adaptive-Grained Scheduling for Global Computing System

Jaesun HAN, Daeyeon PARK

Full Text Views

0

Cite this

Summary :

Global computing system (GCS) harnesses the idle CPU resources of clients connected to Internet for solving large problems that require high volume of computing power. Since GCS scale to millions of clients, many projects usually adopt coarse-grained scheduling in order to reduce server-side contention at the expense of sacrificing the degree of parallelism and wasting CPU resources. In this paper, we propose a new type of client, i.e., a scheduling proxy that enables adaptive-grained scheduling between the server and clients. While the server allocates coarse-grained work units to scheduling proxies alone, clients download fine-grained work units from a relatively nearby scheduling proxy not from the distant server. By computation of small work units at client side, the turnaround time of work unit can be reduced and the waste of CPU time by timeout can be minimized without increasing the performance cost of contention at the server. In addition, in order not to lose results in the failure of scheduling proxies, we suggest a technique of result caching in clients.

Publication: IEICE TRANSACTIONS on Communications Vol.E88-B No.4 pp.1448-1457

Publication Date: 2005/04/01

Publicized

Online ISSN

DOI: 10.1093/ietcom/e88-b.4.1448

Type of Manuscript: Special Section PAPER (Special Section on Internet Technology V)

Category

Cite this

Copy

Jaesun HAN, Daeyeon PARK, "Scheduling Proxy: Enabling Adaptive-Grained Scheduling for Global Computing System" in IEICE TRANSACTIONS on Communications, vol. E88-B, no. 4, pp. 1448-1457, April 2005, doi: 10.1093/ietcom/e88-b.4.1448.
Abstract: Global computing system (GCS) harnesses the idle CPU resources of clients connected to Internet for solving large problems that require high volume of computing power. Since GCS scale to millions of clients, many projects usually adopt coarse-grained scheduling in order to reduce server-side contention at the expense of sacrificing the degree of parallelism and wasting CPU resources. In this paper, we propose a new type of client, i.e., a scheduling proxy that enables adaptive-grained scheduling between the server and clients. While the server allocates coarse-grained work units to scheduling proxies alone, clients download fine-grained work units from a relatively nearby scheduling proxy not from the distant server. By computation of small work units at client side, the turnaround time of work unit can be reduced and the waste of CPU time by timeout can be minimized without increasing the performance cost of contention at the server. In addition, in order not to lose results in the failure of scheduling proxies, we suggest a technique of result caching in clients.
URL: https://global.ieice.org/en_transactions/communications/10.1093/ietcom/e88-b.4.1448/_p

Copy

@ARTICLE{e88-b_4_1448,
author={Jaesun HAN, Daeyeon PARK, },
journal={IEICE TRANSACTIONS on Communications},
title={Scheduling Proxy: Enabling Adaptive-Grained Scheduling for Global Computing System},
year={2005},
volume={E88-B},
number={4},
pages={1448-1457},
abstract={Global computing system (GCS) harnesses the idle CPU resources of clients connected to Internet for solving large problems that require high volume of computing power. Since GCS scale to millions of clients, many projects usually adopt coarse-grained scheduling in order to reduce server-side contention at the expense of sacrificing the degree of parallelism and wasting CPU resources. In this paper, we propose a new type of client, i.e., a scheduling proxy that enables adaptive-grained scheduling between the server and clients. While the server allocates coarse-grained work units to scheduling proxies alone, clients download fine-grained work units from a relatively nearby scheduling proxy not from the distant server. By computation of small work units at client side, the turnaround time of work unit can be reduced and the waste of CPU time by timeout can be minimized without increasing the performance cost of contention at the server. In addition, in order not to lose results in the failure of scheduling proxies, we suggest a technique of result caching in clients.},
keywords={},
doi={10.1093/ietcom/e88-b.4.1448},
ISSN={},
month={April},}

Copy

TY - JOUR
TI - Scheduling Proxy: Enabling Adaptive-Grained Scheduling for Global Computing System
T2 - IEICE TRANSACTIONS on Communications
SP - 1448
EP - 1457
AU - Jaesun HAN
AU - Daeyeon PARK
PY - 2005
DO - 10.1093/ietcom/e88-b.4.1448
JO - IEICE TRANSACTIONS on Communications
SN -
VL - E88-B
IS - 4
JA - IEICE TRANSACTIONS on Communications
Y1 - April 2005
AB - Global computing system (GCS) harnesses the idle CPU resources of clients connected to Internet for solving large problems that require high volume of computing power. Since GCS scale to millions of clients, many projects usually adopt coarse-grained scheduling in order to reduce server-side contention at the expense of sacrificing the degree of parallelism and wasting CPU resources. In this paper, we propose a new type of client, i.e., a scheduling proxy that enables adaptive-grained scheduling between the server and clients. While the server allocates coarse-grained work units to scheduling proxies alone, clients download fine-grained work units from a relatively nearby scheduling proxy not from the distant server. By computation of small work units at client side, the turnaround time of work unit can be reduced and the waste of CPU time by timeout can be minimized without increasing the performance cost of contention at the server. In addition, in order not to lose results in the failure of scheduling proxies, we suggest a technique of result caching in clients.
ER -