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

In a modern DBMS, working memory is frequently the limiting factor whenprocessing in-memory analytic query operations such as joins, sorting, andaggregation. Existing resource estimation approaches for a DBMS estimate theresource consumption of a query by computing an estimate of each individualdatabase operator in the query execution plan. Such an approach is slow anderror-prone as it relies upon simplifying assumptions, such as uniformity andindependence of the underlying data. Additionally, the existing approachfocuses on individual queries separately and does not factor in other queriesin the workload that may be executed concurrently. In this research, we areinterested in query performance optimization under concurrent execution of abatch of queries (a workload). Specifically, we focus on predicting the memorydemand for a workload rather than providing separate estimates for each querywithin it. We introduce the problem of workload memory prediction and formalizeit as a distribution regression problem. We propose Learned Workload MemoryPrediction (LearnedWMP) to improve and simplify estimating the working memorydemands of workloads. Through a comprehensive experimental evaluation, we showthat LearnedWMP reduces the memory estimation error of thestate-of-the-practice method by up to 47.6%. Compared to an alternativesingle-query model, during training and inferencing, the LearnedWMP model andits variants were 3x to 10x faster. Moreover, LearnedWMP-based models were atleast 50% smaller in most cases. Overall, the results demonstrate theadvantages of the LearnedWMP approach and its potential for a broader impact onquery performance optimization.

LearnedWMP: Workload Memory Prediction Using Distribution of Query Templates