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

Parallel aggregation is a ubiquitous operation in data analytics that isexpressed as GROUP BY in SQL, reduce in Hadoop, or segment in TensorFlow.Parallel aggregation starts with an optional local pre-aggregation step andthen repartitions the intermediate result across the network. While localpre-aggregation works well for low-cardinality aggregations, the networkcommunication cost remains significant for high-cardinality aggregations evenafter local pre-aggregation. The problem is that the repartition-basedalgorithm for high-cardinality aggregation does not fully utilize the network.  In this work, we first formulate a mathematical model that captures theperformance of parallel aggregation. We prove that finding optimal aggregationplans from a known data distribution is NP-hard, assuming the Small SetExpansion conjecture. We propose GRASP, a GReedy Aggregation SchedulingProtocol that decomposes parallel aggregation into phases. GRASP isdistribution-aware as it aggregates the most similar partitions in each phaseto reduce the transmitted data size in subsequent phases. In addition, GRASPtakes the available network bandwidth into account when scheduling aggregationsin each phase to maximize network utilization. The experimental evaluation onreal data shows that GRASP outperforms repartition-based aggregation by 3.5xand LOOM by 2.0x.

Chasing Similarity: Distribution-aware Aggregation Scheduling (Extended Version)