Automatic Array Transformation to Columnar Storage at Run Time

Makor, Lukas; Kloibhofer, Sebastian; Leopoldseder, David; Bonetta, Daniele; Mössenböck, Hanspeter

doi:10.1145/3546918.3546919

Cited by 2 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Overall, this work contributes the following: 1. Novel compiler optimizations for the approach proposed in [55] that automatically duplicate the control flow for accessing columnar arrays to mitigate recompilation efforts and to speed up smaller workloads (Section 4). 2.…”

Section: :2mentioning

confidence: 99%

“…Makor et al [55] showed that data-intensive queries on analytical workloads can be significantly accelerated by transforming arrays of uniform objects to columnar arrays and optimizing accesses to the resulting columnar storage accordingly. Such a representation typically improves cache-utilization when iterating over the array and furthermore allows optimization of array accesses in loops at compile time as shown in [55]. To perform such a transformation, information about arrays of objects is tracked at run time.…”

Section: Columnar Arrays In Graalvm Javascriptmentioning

confidence: 99%

“…Each object in the original array is transformed to a proxy object. As stated in [55], the name proxy is adopted from related work [56,71] and is not to be confused with ECMAScript proxy objects [33]. The proxy transformation is necessary, because after the array transformation, each object has to access the columnar storage when it needs to access its data.…”

Section: Columnar Arrays In Graalvm Javascriptmentioning

confidence: 99%

“…The results of Makor et al [55] show significant speedups when executing queries that process non-object-type properties, e.g., boolean or number. However, the results also show that when working with properties of reference types, e.g., dates, strings, or user object types, for which the actual data is stored in properties of these objects, the effect is reduced or performance even degrades.…”

Section: Multi-level Storage Transformationmentioning

confidence: 99%

“…Based on these observations, Makor et al [55] developed an approach for automated transformation of arrays of objects to columnar storage in the GraalVM JavaScript runtime [66]. They detect suitable arrays at run time and transform the underlying storage to a columnar layout if the array is large enough and frequently accessed.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Control Flow Duplication for Columnar Arrays in a Dynamic Compiler

et al. 2023

View full text Add to dashboard Cite

Columnar databases are an established way to speed up online analytical processing (OLAP) queries. Nowadays, data processing (e.g., storage, visualization, and analytics) is often performed at the programming language level, hence it is desirable to also adopt columnar data structures for common language runtimes.While there are frameworks, libraries, and APIs to enable columnar data stores in programming languages, their integration into applications typically requires developer interference. In prior work, researchers implemented an approach for automated transformation of arrays into columnar arrays in the GraalVM JavaScript runtime. However, this approach suffers from performance issues on smaller workloads as well as on more complex nested data structures. We find that the key to optimizing accesses to columnar arrays is to identify queries and apply specific optimizations to them.In this paper, we describe novel compiler optimizations in the GraalVM Compiler that optimize queries on columnar arrays. At JIT compile time, we identify loops that access potentially columnar arrays and duplicate them in order to specifically optimize accesses to columnar arrays. Additionally, we describe a new approach for creating columnar arrays from arrays consisting of complex objects by performing multi-level storage transformation. We demonstrate our approach via an implementation for JavaScript Date objects.Our work shows that automatic transformation of arrays to columnar storage is feasible even for small workloads and that more complex arrays of objects could benefit from a multi-level transformation. Furthermore, we show how we can optimize methods that handle arrays in different states by the use of duplication. We evaluated our work on microbenchmarks and established data analytics workloads (TPC-H) to demonstrate that it significantly outperforms previous efforts, with speedups of up to 10x for particular queries. Queries additionally benefit from multi-level transformation, reaching speedups of around 2x. Additionally, we show that we do not cause significant overhead on workloads not suitable for storage transformation.We argue that automatically created columnar arrays could aid developers in data-centric applications as an alternative approach to using dedicated APIs on manually created columnar arrays. Via automatic detection and optimization of queries on potentially columnar arrays, we can improve performance of data processing and further enable its use in common-particularly dynamic-programming languages. ACM CCS 2012Software and its engineering → Dynamic compilers; Runtime environments; Interpreters; Information systems → Column based storage;

show abstract

Section: :2mentioning

confidence: 99%

Section: Columnar Arrays In Graalvm Javascriptmentioning

confidence: 99%