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

Declarative large-scale machine learning (ML) aims at the specification of MLalgorithms in a high-level language and automatic generation of hybrid runtimeexecution plans ranging from single node, in-memory computations to distributedcomputations on MapReduce (MR) or similar frameworks like Spark. Thecompilation of large-scale ML programs exhibits many opportunities forautomatic optimization. Advanced cost-based optimization techniquesrequire---as a fundamental precondition---an accurate cost model for evaluatingthe impact of optimization decisions. In this paper, we share insights into asimple and robust yet accurate technique for costing alternative runtimeexecution plans of ML programs. Our cost model relies on generating and costingruntime plans in order to automatically reflect all successive optimizationphases. Costing runtime plans also captures control flow structures such asloops and branches, and a variety of cost factors like IO, latency, andcomputation costs. Finally, we linearize all these cost factors into a singlemeasure of expected execution time. Within SystemML, this cost model isleveraged by several advanced optimizers like resource optimization and globaldata flow optimization. We share our lessons learned in order to providefoundations for the optimization of ML programs.

Costing Generated Runtime Execution Plans for Large-Scale Machine Learning Programs