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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Our paper presents solutions that can significantly improve the delayperformance of putting and retrieving data in and out of cloud storage. Wefirst focus on measuring the delay performance of a very popular cloud storageservice Amazon S3. We establish that there is significant randomness in servicetimes for reading and writing small and medium size objects when assigneddistinct keys. We further demonstrate that using erasure coding, parallelconnections to storage cloud and limited chunking (i.e., dividing the objectinto a few smaller objects) together pushes the envelope on service timedistributions significantly (e.g., 76%, 80%, and 85% reductions in mean, 90th,and 99th percentiles for 2 Mbyte files) at the expense of additional storage(e.g., 1.75x). However, chunking and erasure coding increase the load and hencethe queuing delays while reducing the supportable rate region in number ofrequests per second per node. Thus, in the second part of our paper we focus onanalyzing the delay performance when chunking, FEC, and parallel connectionsare used together. Based on this analysis, we develop load adaptive algorithmsthat can pick the best code rate on a per request basis by using off-linecomputed queue backlog thresholds. The solutions work with homogeneous serviceswith fixed object sizes, chunk sizes, operation type (e.g., read or write) aswell as heterogeneous services with mixture of object sizes, chunk sizes, andoperation types. We also present a simple greedy solution thatopportunistically uses idle connections and picks the erasure coding rateaccordingly on the fly. Both backlog and greedy solutions support the full rateregion and provide best mean delay performance when compared to the best fixedcoding rate policy. Our evaluations show that backlog based solutions achievebetter delay performance at higher percentile values than the greedy solution.

FAST CLOUD: Pushing the Envelope on Delay Performance of Cloud Storage with Coding