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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Nowadays, the efficiency and even the feasibility of traditionalload-balancing policies are challenged by the rapid growth of cloudinfrastructure and the increasing levels of server heterogeneity. In suchheterogeneous systems with many load-balancers, traditional solutions, such asJSQ, incur a prohibitively large communication overhead and detrimental incasteffects due to herd behavior. Alternative low-communication policies, such asJSQ(d) and the recently proposed JIQ, are either unstable or provide poorperformance.  We introduce the Local Shortest Queue (LSQ) family of load balancingalgorithms. In these algorithms, each dispatcher maintains its own, local, andpossibly outdated view of the server queue lengths, and keeps using JSQ on itslocal view. A small communication overhead is used infrequently to update thislocal view. We formally prove that as long as the error in these localestimates of the server queue lengths is bounded in expectation, the entiresystem is strongly stable. Finally, in simulations, we show how simple andstable LSQ policies exhibit appealing performance and significantly outperformexisting low-communication policies, while using an equivalent communicationbudget. In particular, our simple policies often outperform even JSQ due totheir reduction of herd behavior. We further show how, by relying on smartservers (i.e., advanced pull-based communication), we can further improveperformance and lower communication overhead.

LSQ: Load Balancing in Large-Scale Heterogeneous Systems with Multiple Dispatchers