In this paper, we propose an extension to GridFTP that optimizes its performance by dynamically adjusting the number of parallel TCP connections. GridFTP has been used as a data transfer protocol to effectively transfer a large volume of data in Grid computing. GridFTP supports a feature called parallel data transfer that improves throughput by establishing multiple TCP connections in parallel. However, for achieving high GridFTP throughput, the number of TCP connections should be optimized based on the network status. In this paper, we propose an automatic parallelism tuning mechanism called GridFTP-APT (GridFTP with Automatic Parallelism Tuning) that adjusts the number of parallel TCP connections according to information available to the Grid middleware. Through simulations, we demonstrate that GridFTP-APT significantly improves the performance of GridFTP in various network environments.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Takeshi ITO, Hiroyuki OHSAKI, Makoto IMASE, "GridFTP-APT: Automatic Parallelism Tuning Mechanism for GridFTP in Long-Fat Networks" in IEICE TRANSACTIONS on Communications,
vol. E91-B, no. 12, pp. 3925-3936, December 2008, doi: 10.1093/ietcom/e91-b.12.3925.
Abstract: In this paper, we propose an extension to GridFTP that optimizes its performance by dynamically adjusting the number of parallel TCP connections. GridFTP has been used as a data transfer protocol to effectively transfer a large volume of data in Grid computing. GridFTP supports a feature called parallel data transfer that improves throughput by establishing multiple TCP connections in parallel. However, for achieving high GridFTP throughput, the number of TCP connections should be optimized based on the network status. In this paper, we propose an automatic parallelism tuning mechanism called GridFTP-APT (GridFTP with Automatic Parallelism Tuning) that adjusts the number of parallel TCP connections according to information available to the Grid middleware. Through simulations, we demonstrate that GridFTP-APT significantly improves the performance of GridFTP in various network environments.
URL: https://global.ieice.org/en_transactions/communications/10.1093/ietcom/e91-b.12.3925/_p
Copy
@ARTICLE{e91-b_12_3925,
author={Takeshi ITO, Hiroyuki OHSAKI, Makoto IMASE, },
journal={IEICE TRANSACTIONS on Communications},
title={GridFTP-APT: Automatic Parallelism Tuning Mechanism for GridFTP in Long-Fat Networks},
year={2008},
volume={E91-B},
number={12},
pages={3925-3936},
abstract={In this paper, we propose an extension to GridFTP that optimizes its performance by dynamically adjusting the number of parallel TCP connections. GridFTP has been used as a data transfer protocol to effectively transfer a large volume of data in Grid computing. GridFTP supports a feature called parallel data transfer that improves throughput by establishing multiple TCP connections in parallel. However, for achieving high GridFTP throughput, the number of TCP connections should be optimized based on the network status. In this paper, we propose an automatic parallelism tuning mechanism called GridFTP-APT (GridFTP with Automatic Parallelism Tuning) that adjusts the number of parallel TCP connections according to information available to the Grid middleware. Through simulations, we demonstrate that GridFTP-APT significantly improves the performance of GridFTP in various network environments.},
keywords={},
doi={10.1093/ietcom/e91-b.12.3925},
ISSN={1745-1345},
month={December},}
Copy
TY - JOUR
TI - GridFTP-APT: Automatic Parallelism Tuning Mechanism for GridFTP in Long-Fat Networks
T2 - IEICE TRANSACTIONS on Communications
SP - 3925
EP - 3936
AU - Takeshi ITO
AU - Hiroyuki OHSAKI
AU - Makoto IMASE
PY - 2008
DO - 10.1093/ietcom/e91-b.12.3925
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E91-B
IS - 12
JA - IEICE TRANSACTIONS on Communications
Y1 - December 2008
AB - In this paper, we propose an extension to GridFTP that optimizes its performance by dynamically adjusting the number of parallel TCP connections. GridFTP has been used as a data transfer protocol to effectively transfer a large volume of data in Grid computing. GridFTP supports a feature called parallel data transfer that improves throughput by establishing multiple TCP connections in parallel. However, for achieving high GridFTP throughput, the number of TCP connections should be optimized based on the network status. In this paper, we propose an automatic parallelism tuning mechanism called GridFTP-APT (GridFTP with Automatic Parallelism Tuning) that adjusts the number of parallel TCP connections according to information available to the Grid middleware. Through simulations, we demonstrate that GridFTP-APT significantly improves the performance of GridFTP in various network environments.
ER -