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/

Zendy Plus

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

Zendy Tools

Zendy Open

Query processing over big data is ubiquitous in modern clouds, where thesystem takes care of picking both the physical query execution plans and theresources needed to run those plans, using a cost-based query optimizer. A goodcost model, therefore, is akin to better resource efficiency and loweroperational costs. Unfortunately, the production workloads at Microsoft showthat costs are very complex to model for big data systems. In this work, weinvestigate two key questions: (i) can we learn accurate cost models for bigdata systems, and (ii) can we integrate the learned models within the queryoptimizer. To answer these, we make three core contributions. First, we exploitworkload patterns to learn a large number of individual cost models and combinethem to achieve high accuracy and coverage over a long period. Second, wepropose extensions to Cascades framework to pick optimal resources, i.e, numberof containers, during query planning. And third, we integrate the learned costmodels within the Cascade-style query optimizer of SCOPE at Microsoft. Weevaluate the resulting system, Cleo, in a production environment using bothproduction and TPC-H workloads. Our results show that the learned cost modelsare 2 to 3 orders of magnitude more accurate, and 20X more correlated with theactual runtimes, with a large majority (70%) of the plan changes leading tosubstantial improvements in latency as well as resource usage.

Cost Models for Big Data Query Processing: Learning, Retrofitting, and Our Findings