Abstract
This paper explores the changes required of TCP to efficiently support cluster file systems such as Hadoop Distributed File System (HDFS) where the storage nodes are connected through a software defined networking (SDN). Traditional chain replications in these file systems incur large delay and cause inefficient network use. But SDN can cooperate with the cluster file systems to address the problems by pre-Arranging a distribution tree, which opens the possibility of parallel replication. Unfortunately, it cannot be realized without extending TCP, to accommodate the parallel transfer on the transport layer. This paper discusses how to extend TCP to make it possible, and demonstrates the feasibility by implementing a prototype in the Linux kernel. The prototype saves the data replication time by 25% while substantially reducing network use.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2019 IEEE Conference on Network Softwarization |
| Subtitle of host publication | Unleashing the Power of Network Softwarization, NetSoft 2019 |
| Editors | Filip De Turck, Flavio Esposito, Prosper Chemouil, Olivier Festor, Stefano Secci, Christian Jacquenet, Walter Cerroni |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 48-53 |
| Number of pages | 6 |
| ISBN (Electronic) | 9781538693766 |
| DOIs | |
| Publication status | Published - 2019 Jun 1 |
| Event | 5th IEEE Conference on Network Softwarization, NetSoft 2019 - Paris, France Duration: 2019 Jun 24 → 2019 Jun 28 |
Publication series
| Name | Proceedings of the 2019 IEEE Conference on Network Softwarization: Unleashing the Power of Network Softwarization, NetSoft 2019 |
|---|
Conference
| Conference | 5th IEEE Conference on Network Softwarization, NetSoft 2019 |
|---|---|
| Country | France |
| City | Paris |
| Period | 19/6/24 → 19/6/28 |
Fingerprint
Keywords
- cluster file system
- data replication
- HDFS
- network utilization
- Software Defined Networking (SDN)
- TCP extension
ASJC Scopus subject areas
- Safety, Risk, Reliability and Quality
- Computer Networks and Communications
- Software
Cite this
Extending TCP for accelerating replication on cluster file systems over sdns. / Lim, Sungheon; Kim, Hyogon.
Proceedings of the 2019 IEEE Conference on Network Softwarization: Unleashing the Power of Network Softwarization, NetSoft 2019. ed. / Filip De Turck; Flavio Esposito; Prosper Chemouil; Olivier Festor; Stefano Secci; Christian Jacquenet; Walter Cerroni. Institute of Electrical and Electronics Engineers Inc., 2019. p. 48-53 8806672 (Proceedings of the 2019 IEEE Conference on Network Softwarization: Unleashing the Power of Network Softwarization, NetSoft 2019).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Extending TCP for accelerating replication on cluster file systems over sdns
AU - Lim, Sungheon
AU - Kim, Hyogon
PY - 2019/6/1
Y1 - 2019/6/1
N2 - This paper explores the changes required of TCP to efficiently support cluster file systems such as Hadoop Distributed File System (HDFS) where the storage nodes are connected through a software defined networking (SDN). Traditional chain replications in these file systems incur large delay and cause inefficient network use. But SDN can cooperate with the cluster file systems to address the problems by pre-Arranging a distribution tree, which opens the possibility of parallel replication. Unfortunately, it cannot be realized without extending TCP, to accommodate the parallel transfer on the transport layer. This paper discusses how to extend TCP to make it possible, and demonstrates the feasibility by implementing a prototype in the Linux kernel. The prototype saves the data replication time by 25% while substantially reducing network use.
AB - This paper explores the changes required of TCP to efficiently support cluster file systems such as Hadoop Distributed File System (HDFS) where the storage nodes are connected through a software defined networking (SDN). Traditional chain replications in these file systems incur large delay and cause inefficient network use. But SDN can cooperate with the cluster file systems to address the problems by pre-Arranging a distribution tree, which opens the possibility of parallel replication. Unfortunately, it cannot be realized without extending TCP, to accommodate the parallel transfer on the transport layer. This paper discusses how to extend TCP to make it possible, and demonstrates the feasibility by implementing a prototype in the Linux kernel. The prototype saves the data replication time by 25% while substantially reducing network use.
KW - cluster file system
KW - data replication
KW - HDFS
KW - network utilization
KW - Software Defined Networking (SDN)
KW - TCP extension
UR - http://www.scopus.com/inward/record.url?scp=85072033690&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85072033690&partnerID=8YFLogxK
U2 - 10.1109/NETSOFT.2019.8806672
DO - 10.1109/NETSOFT.2019.8806672
M3 - Conference contribution
AN - SCOPUS:85072033690
T3 - Proceedings of the 2019 IEEE Conference on Network Softwarization: Unleashing the Power of Network Softwarization, NetSoft 2019
SP - 48
EP - 53
BT - Proceedings of the 2019 IEEE Conference on Network Softwarization
A2 - De Turck, Filip
A2 - Esposito, Flavio
A2 - Chemouil, Prosper
A2 - Festor, Olivier
A2 - Secci, Stefano
A2 - Jacquenet, Christian
A2 - Cerroni, Walter
PB - Institute of Electrical and Electronics Engineers Inc.
ER -