Managing Response Time Tails by Sharding
Managing Response Time Tails by Sharding
复制标题
通过分片管理响应时间尾部
DOI:
10.1145/3300143
复制
发表时间:
2019
影响因子:
0.6
通讯作者:
Harrison P
中科院分区:
文献类型:
--
作者:
Harrison P
Matrix analytic methods are developed to compute the probability distribution of response times (i.e., data access times) in distributed storage systems protected by erasure coding, which is implemented by sharding a data object intoNfragments, onlyK<;Nof which are required to reconstruct the object. This leads to a partial-fork-join model with a choice of canceling policies for the redundantN−Ktasks. The accuracy of the analytical model is supported by tests against simulation in a broad range of setups. At increasing workload intensities, numerical results show the extent to which increasing the redundancy level reduces the mean response time of storage reads and significantly flattens the tail of their distribution; this is demonstrated at medium-high quantiles, up to the 99th. The quantitative reduction in response time achieved by two policies for canceling redundant tasks is also shown: for cancel-at-finish and cancel-at-start, which limits the additional load introduced whilst losing the benefit of selectivity amongst fragment service times.
登录
查看更多内容
DOI:
--
发表时间:
2006
期刊:
European Performance Engineering Workshop
影响因子:
--
作者:
A. Heindl;G. Horváth;K. Gross
通讯作者:
K. Gross
DOI:
10.1109/tnet.2015.2466453
发表时间:
2014-04
期刊:
IEEE/ACM Transactions on Networking
影响因子:
--
作者:
Yu Xiang;Tian Lan;V. Aggarwal;Y. Chen
通讯作者:
Yu Xiang;Tian Lan;V. Aggarwal;Y. Chen
DOI:
10.1145/3159172
发表时间:
2018
影响因子:
0.6
作者:
Li, Bin;Ramamoorthy, Aditya;Srikant, R.
通讯作者:
Srikant, R.
影响因子:
2.1
作者:
B. V. Houdt;J. V. Leeuwaarden
通讯作者:
J. V. Leeuwaarden
DOI:
--
发表时间:
2012
期刊:
影响因子:
--
作者:
Nihar B. Shah;Kangwook Lee;K. Ramchandran
通讯作者:
K. Ramchandran