Implicit parallelism through deep language embedding

Alexander Alexandrov, Andreas Kunft, Asterios Katsifodimos, Felix Schüler, Lauritz Thamsen, Odej Kao, Tobias Herb, Volker Markl

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

27 Citations (Scopus)

Abstract

The appeal of MapReduce has spawned a family of systems that implement or extend it. In order to enable parallel collection processing with User-Defined Functions (UDFs), these systems expose extensions of the MapReduce programming model as library-based dataow APIs that are tightly coupled to their underlying runtime engine. Expressing data analysis algorithms with complex data and control ow structure using such APIs reveals a number of limitations that impede programmer's productivity. In this paper we show that the design of data analysis languages and APIs from a runtime engine point of view bloats the APIs with low-level primitives and affects programmer's productivity. Instead, we argue that an approach based on deeply embedding the APIs in a host language can address the shortcomings of current data analysis languages. To demonstrate this, we propose a language for complex data analysis embedded in Scala, which (i) allows for declarative specification of dataows and (ii) hides the notion of dataparallelism and distributed runtime behind a suitable intermediate representation. We describe a compiler pipeline that facilitates efficient data-parallel processing without imposing runtime engine-bound syntactic or semantic restrictions on the structure of the input programs. We present a series of experiments with two state-of-the-art systems that demonstrate the optimization potential of our approach.

Original languageEnglish
Title of host publicationSIGMOD'15
Subtitle of host publicationProceedings of the 2015 ACM SIGMOD International Conference on Management of Data
PublisherAssociation for Computing Machinery (ACM)
Pages47-61
Number of pages15
ISBN (Print)978-1-4503-2758-9
DOIs
Publication statusPublished - 27 May 2015
Externally publishedYes
EventACM SIGMOD International Conference on Management of Data, SIGMOD 2015 - Melbourne, Australia
Duration: 31 May 20154 Jun 2015

Conference

ConferenceACM SIGMOD International Conference on Management of Data, SIGMOD 2015
CountryAustralia
CityMelbourne
Period31/05/154/06/15

Fingerprint Dive into the research topics of 'Implicit parallelism through deep language embedding'. Together they form a unique fingerprint.

Cite this