CompGen: generation of fast JIT compilers in a multi-language VM

Latifi, Florian; Leopoldseder, David; Wimmer, Christian; Mössenböck, Hanspeter

doi:10.1145/3486602.3486930

Cited by 5 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Truffle [39,85,88] is a language implementation framework that enables guestlanguage integration via AST interpretation. At run time, the guest-language ASTs are partially evaluated and optimized based on profiling information accumulated during interpretation [42,50,87,89] and are subsequently compiled using the GraalVM Compiler. Truffle furthermore provides a standardized object storage model [86] that can be used across all language implementations.…”

Section: Trufflementioning

confidence: 99%

See 1 more Smart Citation

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: Trufflementioning

confidence: 99%

“…The loop must not contain method calls or other nodes which can allow the array to escape (and potentially change its state). While this may appear limiting, the GraalVM partial evaluator inlines most AST operations and nested calls, such that they should only appear infrequently within loop bodies [42,50,87,89].…”

Section: Loop Analysis and Selectionmentioning

confidence: 99%

Control Flow Duplication for Columnar Arrays in a Dynamic Compiler

et al. 2023

View full text Add to dashboard Cite

show abstract

Automatic Array Transformation to Columnar Storage at Run Time

Makor

Kloibhofer

Leopoldseder³

et al. 2022

Proceedings of the 19th International Conference on Managed Programming Languages and Runtimes

View full text Add to dashboard Cite

Specialization of interpreters written in object-oriented languages can be effective

Adamovich¹,

Klimov²

2022

Proceedings of 24th Scientific Conference “Scientific Services &Amp; Internet – 2022”

View full text Add to dashboard Cite

Barriers of real object-oriented program specialization can be often overcome using modern metacomputation techniques. One of the barriers is the resolution of polymorphism at the stage of program analysis before the 4 execution of the program. The last problem is successfully solved for a number of cases in the JaSpe specializer, as shown in this paper. The paper is devoted to the program compilation by specialization methods, without the use of a compiler. We have applied the partial evaluator JaSpe to two arithmetic expression language interpreters written in Java. The interpreters were implemented using the recursive descent method and the visitor pattern. As a result of the successful specialization of these interpreters by the square root program written on arithmetic expression language, compiled versions of the latter were obtained. In this case, the acceleration was from 12 to 22 times.

show abstract

CompGen: generation of fast JIT compilers in a multi-language VM

Cited by 5 publications

References 17 publications

Control Flow Duplication for Columnar Arrays in a Dynamic Compiler

Control Flow Duplication for Columnar Arrays in a Dynamic Compiler

Automatic Array Transformation to Columnar Storage at Run Time

Specialization of interpreters written in object-oriented languages can be effective

Contact Info

Product

Resources

About