Fast Link Failure Detection in Datacenter
碩士 === 國立中央大學 === 資訊工程學系 === 105 === Load balancing is an important technique to cope with dynamic and unpredictable traffic demands in data center networks. In general, load balancing schemes aim to split traffics evenly among multiple paths. However, most existing approaches either suffers from pa...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2017
|
Online Access: | http://ndltd.ncl.edu.tw/handle/69299401639114331025 |
id |
ndltd-TW-105NCU05392090 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-105NCU053920902017-10-22T04:29:53Z http://ndltd.ncl.edu.tw/handle/69299401639114331025 Fast Link Failure Detection in Datacenter 在資料中心的快速線路異常偵測 Wun-Sin Wu 吳文心 碩士 國立中央大學 資訊工程學系 105 Load balancing is an important technique to cope with dynamic and unpredictable traffic demands in data center networks. In general, load balancing schemes aim to split traffics evenly among multiple paths. However, most existing approaches either suffers from packet reordering (which may confuse TCP congestion control) or fail to quick response (i.e., coarse slicing granularity). Recently, FLARE introduced a burst (called flowlet) based traffic splitting, which attains responsiveness without causing packet reordering. However, the very high bandwidth of internal datacenter flows suggests that the gaps needed for flowlets may be rare. Besides, in Flare, splitting granularity increases (i.e., coarse granularity) when flow size increases. In this paper, we propose an artificial flowlet-based load balancing algorithm which can maintain fine-granularity (even in large flows) and can also avoid packet reordering. Our scheme has at least 20% improvement in flow completion time under the same incidence of packet reordering. Guey-Yun Chang 張貴雲 2017 學位論文 ; thesis 59 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中央大學 === 資訊工程學系 === 105 === Load balancing is an important technique to cope with dynamic and unpredictable
traffic demands in data center networks. In general, load balancing
schemes aim to split traffics evenly among multiple paths. However,
most existing approaches either suffers from packet reordering (which may
confuse TCP congestion control) or fail to quick response (i.e., coarse slicing
granularity). Recently, FLARE introduced a burst (called flowlet) based traffic
splitting, which attains responsiveness without causing packet reordering.
However, the very high bandwidth of internal datacenter flows suggests that
the gaps needed for flowlets may be rare. Besides, in Flare, splitting granularity
increases (i.e., coarse granularity) when flow size increases. In this
paper, we propose an artificial flowlet-based load balancing algorithm which
can maintain fine-granularity (even in large flows) and can also avoid packet
reordering. Our scheme has at least 20% improvement in flow completion
time under the same incidence of packet reordering.
|
author2 |
Guey-Yun Chang |
author_facet |
Guey-Yun Chang Wun-Sin Wu 吳文心 |
author |
Wun-Sin Wu 吳文心 |
spellingShingle |
Wun-Sin Wu 吳文心 Fast Link Failure Detection in Datacenter |
author_sort |
Wun-Sin Wu |
title |
Fast Link Failure Detection in Datacenter |
title_short |
Fast Link Failure Detection in Datacenter |
title_full |
Fast Link Failure Detection in Datacenter |
title_fullStr |
Fast Link Failure Detection in Datacenter |
title_full_unstemmed |
Fast Link Failure Detection in Datacenter |
title_sort |
fast link failure detection in datacenter |
publishDate |
2017 |
url |
http://ndltd.ncl.edu.tw/handle/69299401639114331025 |
work_keys_str_mv |
AT wunsinwu fastlinkfailuredetectionindatacenter AT wúwénxīn fastlinkfailuredetectionindatacenter AT wunsinwu zàizīliàozhōngxīndekuàisùxiànlùyìchángzhēncè AT wúwénxīn zàizīliàozhōngxīndekuàisùxiànlùyìchángzhēncè |
_version_ |
1718556545351942144 |