Tuning GridFTP Pipelining, Concurrency and Parallelism Based on Historical Data

Jangyoung KIM

doi:10.1587/transinf.2014EDL8104

Tuning GridFTP Pipelining, Concurrency and Parallelism Based on Historical Data

Jangyoung KIM

Full Text Views

0

Share
Cite this

Summary :

This paper presents a prediction model based on historical data to achieve optimal values of pipelining, concurrency and parallelism (PCP) in GridFTP data transfers in Cloud systems. Setting the correct values for these three parameters is crucial in achieving high throughput in end-to-end data movement. However, predicting and setting the optimal values for these parameters is a challenging task, especially in shared and non-predictive network conditions. Several factors can affect the optimal values for these parameters such as the background network traffic, available bandwidth, Round-Trip Time (RTT), TCP buffer size, and file size. Existing models either fail to provide accurate predictions or come with very high prediction overheads. The author shows that new model based on historical data can achieve high accuracy with low overhead.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.11 pp.2963-2966

Publication Date: 2014/11/01

Publicized: 2014/07/28

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2014EDL8104

Type of Manuscript: LETTER

Category: Information Network

Authors

Jangyoung KIM
University of Suwon

Keyword

big data, throughput optimization, throughput estimation, pipelining, concurrency, parallelism

Cite this

Copy

Jangyoung KIM, "Tuning GridFTP Pipelining, Concurrency and Parallelism Based on Historical Data" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 11, pp. 2963-2966, November 2014, doi: 10.1587/transinf.2014EDL8104.
Abstract: This paper presents a prediction model based on historical data to achieve optimal values of pipelining, concurrency and parallelism (PCP) in GridFTP data transfers in Cloud systems. Setting the correct values for these three parameters is crucial in achieving high throughput in end-to-end data movement. However, predicting and setting the optimal values for these parameters is a challenging task, especially in shared and non-predictive network conditions. Several factors can affect the optimal values for these parameters such as the background network traffic, available bandwidth, Round-Trip Time (RTT), TCP buffer size, and file size. Existing models either fail to provide accurate predictions or come with very high prediction overheads. The author shows that new model based on historical data can achieve high accuracy with low overhead.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2014EDL8104/_p

Copy

@ARTICLE{e97-d_11_2963,
author={Jangyoung KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Tuning GridFTP Pipelining, Concurrency and Parallelism Based on Historical Data},
year={2014},
volume={E97-D},
number={11},
pages={2963-2966},
abstract={This paper presents a prediction model based on historical data to achieve optimal values of pipelining, concurrency and parallelism (PCP) in GridFTP data transfers in Cloud systems. Setting the correct values for these three parameters is crucial in achieving high throughput in end-to-end data movement. However, predicting and setting the optimal values for these parameters is a challenging task, especially in shared and non-predictive network conditions. Several factors can affect the optimal values for these parameters such as the background network traffic, available bandwidth, Round-Trip Time (RTT), TCP buffer size, and file size. Existing models either fail to provide accurate predictions or come with very high prediction overheads. The author shows that new model based on historical data can achieve high accuracy with low overhead.},
keywords={},
doi={10.1587/transinf.2014EDL8104},
ISSN={1745-1361},
month={November},}

Copy

TY - JOUR
TI - Tuning GridFTP Pipelining, Concurrency and Parallelism Based on Historical Data
T2 - IEICE TRANSACTIONS on Information
SP - 2963
EP - 2966
AU - Jangyoung KIM
PY - 2014
DO - 10.1587/transinf.2014EDL8104
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2014
AB - This paper presents a prediction model based on historical data to achieve optimal values of pipelining, concurrency and parallelism (PCP) in GridFTP data transfers in Cloud systems. Setting the correct values for these three parameters is crucial in achieving high throughput in end-to-end data movement. However, predicting and setting the optimal values for these parameters is a challenging task, especially in shared and non-predictive network conditions. Several factors can affect the optimal values for these parameters such as the background network traffic, available bandwidth, Round-Trip Time (RTT), TCP buffer size, and file size. Existing models either fail to provide accurate predictions or come with very high prediction overheads. The author shows that new model based on historical data can achieve high accuracy with low overhead.
ER -