Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters
Bandwidth guarantee is a critical feature to enable performance predictability in cloud datacenters. This process is expected to achieve three requirements: work conservation, fairness, and simplicity. However, the distributed nature of datacenters raises significant challenges to attaining those re...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8772053/ |
id |
doaj-77aa9008fa184425b8d16114f09e1153 |
---|---|
record_format |
Article |
spelling |
doaj-77aa9008fa184425b8d16114f09e11532021-04-05T17:03:24ZengIEEEIEEE Access2169-35362019-01-01710913410915010.1109/ACCESS.2019.29308888772053Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud DatacentersBaraa Saeed Ali0Kang Chen1https://orcid.org/0000-0002-2376-2898Imran Khan2Department of Electrical and Computer Engineering, Wayne State University, Detroit, MI, USADepartment of Electrical and Computer Engineering, Southern Illinois University, Carbondale, IL, USADepartment of Electrical and Computer Engineering, Southern Illinois University, Carbondale, IL, USABandwidth guarantee is a critical feature to enable performance predictability in cloud datacenters. This process is expected to achieve three requirements: work conservation, fairness, and simplicity. However, the distributed nature of datacenters raises significant challenges to attaining those requirements at the same time. In this paper, we propose an efficient approach that can satisfy the three requirements simultaneously. Our scheme takes advantage of multipath TCP (MPTCP) to generate explicit bandwidth guarantee (BG) traffic and work conservation (WC) traffic. We further prioritize the BG traffic over the WC traffic in the network fabric. Due to the priority setting, WC cannot harm bandwidth guarantees and thus is effectively supported. We show that the MPTCP fits this direction well but presents some new issues when the WC subflows own a low priority. We thus adapt the MPTCP to handle these issues through a customized scheduler (which strictly prioritizes BG subflow during packet scheduling) and adopting a large receive buffer. In addition, we enable tenants to share unused bandwidth fairly by managing the overall aggressiveness of the WC traffic. The proposed system can be easily implemented with commercial off-the-shelf servers and switches. We have implemented with the Linux kernel MPTCP for experiments. The extensive experiments in a small cluster (including one MapReduce experiment) and trace-driven simulations show that our scheme achieves the design goals effectively.https://ieeexplore.ieee.org/document/8772053/Datacenter networkbandwidth guaranteework conservationMPTCPfairness |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Baraa Saeed Ali Kang Chen Imran Khan |
spellingShingle |
Baraa Saeed Ali Kang Chen Imran Khan Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters IEEE Access Datacenter network bandwidth guarantee work conservation MPTCP fairness |
author_facet |
Baraa Saeed Ali Kang Chen Imran Khan |
author_sort |
Baraa Saeed Ali |
title |
Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters |
title_short |
Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters |
title_full |
Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters |
title_fullStr |
Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters |
title_full_unstemmed |
Towards Efficient, Work-Conserving, and Fair Bandwidth Guarantee in Cloud Datacenters |
title_sort |
towards efficient, work-conserving, and fair bandwidth guarantee in cloud datacenters |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
Bandwidth guarantee is a critical feature to enable performance predictability in cloud datacenters. This process is expected to achieve three requirements: work conservation, fairness, and simplicity. However, the distributed nature of datacenters raises significant challenges to attaining those requirements at the same time. In this paper, we propose an efficient approach that can satisfy the three requirements simultaneously. Our scheme takes advantage of multipath TCP (MPTCP) to generate explicit bandwidth guarantee (BG) traffic and work conservation (WC) traffic. We further prioritize the BG traffic over the WC traffic in the network fabric. Due to the priority setting, WC cannot harm bandwidth guarantees and thus is effectively supported. We show that the MPTCP fits this direction well but presents some new issues when the WC subflows own a low priority. We thus adapt the MPTCP to handle these issues through a customized scheduler (which strictly prioritizes BG subflow during packet scheduling) and adopting a large receive buffer. In addition, we enable tenants to share unused bandwidth fairly by managing the overall aggressiveness of the WC traffic. The proposed system can be easily implemented with commercial off-the-shelf servers and switches. We have implemented with the Linux kernel MPTCP for experiments. The extensive experiments in a small cluster (including one MapReduce experiment) and trace-driven simulations show that our scheme achieves the design goals effectively. |
topic |
Datacenter network bandwidth guarantee work conservation MPTCP fairness |
url |
https://ieeexplore.ieee.org/document/8772053/ |
work_keys_str_mv |
AT baraasaeedali towardsefficientworkconservingandfairbandwidthguaranteeinclouddatacenters AT kangchen towardsefficientworkconservingandfairbandwidthguaranteeinclouddatacenters AT imrankhan towardsefficientworkconservingandfairbandwidthguaranteeinclouddatacenters |
_version_ |
1721540369657626624 |