TY - GEN
T1 - An implementation of job migration function using CRIU and podman in docker-based user-PC computing system
AU - Htet, Hein
AU - Funabiki, Nobuo
AU - Kamoyedji, Ariel
AU - Zhou, Xudong
AU - Kuribayashi, Minoru
N1 - Publisher Copyright:
© 2021 Association for Computing Machinery.
PY - 2021
Y1 - 2021
N2 - To offer a low-cost and high-performance distributed computing platform, we have studied the user-PC computing (UPC) system based on the master-worker model. The UPC system uses idling resources of personal computers (PCs) for daily usage by users as the workers, to run the requested jobs or application programs that may need various environments on Docker containers. In this paper, we implement a job migration function in the UPC system to speed up the completion by dynamically changing the assigned worker. It adopts Checkpoint-Restore in Userspace (CRIU) to save the data at the job running into image files and Podman to manage the Docker containers. To verify the function, we conduct extensive measurements with nine jobs and four PCs that have different features. The results show that any job was successfully migrated between different PCs, and the migration from a slow PC to a faster PC reduced the total CPU time.
AB - To offer a low-cost and high-performance distributed computing platform, we have studied the user-PC computing (UPC) system based on the master-worker model. The UPC system uses idling resources of personal computers (PCs) for daily usage by users as the workers, to run the requested jobs or application programs that may need various environments on Docker containers. In this paper, we implement a job migration function in the UPC system to speed up the completion by dynamically changing the assigned worker. It adopts Checkpoint-Restore in Userspace (CRIU) to save the data at the job running into image files and Podman to manage the Docker containers. To verify the function, we conduct extensive measurements with nine jobs and four PCs that have different features. The results show that any job was successfully migrated between different PCs, and the migration from a slow PC to a faster PC reduced the total CPU time.
KW - CRIU
KW - Docker
KW - Job migration
KW - Podman
KW - UPC system
UR - http://www.scopus.com/inward/record.url?scp=85121319057&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85121319057&partnerID=8YFLogxK
U2 - 10.1145/3479162.3479176
DO - 10.1145/3479162.3479176
M3 - Conference contribution
AN - SCOPUS:85121319057
T3 - ACM International Conference Proceeding Series
SP - 92
EP - 97
BT - ICCCM 2021 - Proceedings of the 9th International Conference on Computer and Communications Management
PB - Association for Computing Machinery
T2 - 9th International Conference on Computer and Communications Management, ICCCM 2021
Y2 - 16 July 2021 through 18 July 2021
ER -