English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This paper proposes an algorithm to minimize weighted service latency fordifferent classes of tenants (or service classes) in a data center networkwhere erasure-coded files are stored on distributed disks/racks and accessrequests are scattered across the network. Due to limited bandwidth availableat both top-of-the-rack and aggregation switches and tenants in differentservice classes need differentiated services, network bandwidth must beapportioned among different intra- and inter-rack data flows for differentservice classes in line with their traffic statistics. We formulate thisproblem as weighted queuing and employ a class of probabilistic requestscheduling policies to derive a closed-form upper-bound of service latency forerasure-coded storage with arbitrary file access patterns and service timedistributions. The result enables us to propose a joint weighted latency (overdifferent service classes) optimization over three entangled "control knobs":the bandwidth allocation at top-of-the-rack and aggregation switches fordifferent service classes, dynamic scheduling of file requests, and theplacement of encoded file chunks (i.e., data locality). The joint optimizationis shown to be a mixed-integer problem. We develop an iterative algorithm whichdecouples and solves the joint optimization as 3 sub-problems, which are eitherconvex or solvable via bipartite matching in polynomial time. The proposedalgorithm is prototyped in an open-source, distributed file system, {\emTahoe}, and evaluated on a cloud testbed with 16 separate physical hosts in anOpenstack cluster using 48-port Cisco Catalyst switches. Experiments validateour theoretical latency analysis and show significant latency reduction fordiverse file access patterns. The results provide valuable insights ondesigning low-latency data center networks with erasure coded storage.

Differentiated latency in data center networks with erasure coded files through traffic engineering