Relative factors in performance analysis of Java virtual machines

Gu, Dayong; Verbrugge, Clark; Gagnon, Étienne

doi:10.1145/1134760.1134776

Cited by 21 publications

(13 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since SableVM did not have an optimizing compiler, it was hard for Gagnon to perform a detailed performance evaluation of this design. More recently Gu, Verbrugge and Gagnon [13] set out to compare the performance of this layout in Jikes RVM but concluded that it was difficult to accurately evaluate such design choices in the context of a complex, non-deterministic JVM. Dayong Gu generously made available to us his port to Jikes RVM of the SableVM bidirectional object model, which we forwardported, tuned, and used in our evaluation of the bidirectional object model reported here.…”

Section: Related Workmentioning

confidence: 99%

A comprehensive evaluation of object scanning techniques

Garner

Blackburn

Frampton

2011

Proceedings of the International Symposium on Memory Management

View full text Add to dashboard Cite

At the heart of all garbage collectors lies the process of identifying and processing reference fields within an object. Despite its key role, and evidence of many different implementation approaches, to our knowledge no comprehensive quantitative study of this design space exists. The lack of such a study means that implementers must rely on 'conventional wisdom', hearsay, and their own costly analysis. Starting with mechanisms described in the literature and a variety of permutations of these, we explore the impact of a number of dimensions including: a) the choice of data structure, b) levels of indirection from object to metadata, and c) specialization of scanning code. We perform a comprehensive examination of these tradeoffs on four different architectures using eighteen benchmarks and hardware performance counters. We inform the choice of mechanism with a detailed study of heap composition and object structure as seen by the garbage collector on these benchmarks. Our results show that choice of scanning mechanism is important. We find that a careful choice of scanning mechanism alone can improve garbage collection performance by 16% and total time by 2.5%, on average, over a well tuned baseline. We observe substantial variation in performance among architectures, and find that some mechanisms-particularly specialization, layout of reference fields in objects, and encoding metadata in object headers-yield consistent, significant advantages.

show abstract

Section: Related Workmentioning

confidence: 99%

A comprehensive evaluation of object scanning techniques

Garner

Blackburn

Frampton

2011

Proceedings of the International Symposium on Memory Management

View full text Add to dashboard Cite

show abstract

“…Hardware information obtained from hardware counters can also be used to improve static compilers; Cavazos et al use performance counters to as a means of determining good compiler optimization settings [10]. Other works based on hardware event information can be found in [32,38,18]. Many software applications and libraries are available to access these counters, including VTune, PMAPI, PCL and PAPI [7].…”

Section: Related Workmentioning

confidence: 99%

“…Although this is only one of many possible choices, Gu and Verbrugge have previously shown the importance of "L1 instruction cache miss" on JVM performance [18]. Of course the L1 I-cache is a global resource, subject to potential pollution by other processes.…”

Section: Hardware Performance Datamentioning

confidence: 99%

Phase-based adaptive recompilation in a JVM

Verbrugge

2008

Proceedings of the 6th Annual IEEE/ACM International Symposium on Code Generation and Optimization

Self Cite

View full text Add to dashboard Cite

Modern JIT compilers often employ multi-level recompilation strategies as a means of ensuring the most used code is also the most highly optimized, balancing optimization costs and expected future performance. Accurate selection of code to compile and level of optimization to apply is thus important to performance. In this paper we investigate the effect of an improved recompilation strategy for a Java virtual machine. Our design makes use of a lightweight, low-level profiling mechanism to detect high-level, variable length phases in program execution. Phases are then used to guide adaptive recompilation choices, improving performance. We develop both an offline implementation based on trace data and a self-contained online version. Our offline study shows an average speedup of 8.7% and up to 21%, and our online system achieves an average speedup of 4.4%, up to 18%. We subject our results to extensive analysis and show that our design achieves good overall performance with high consistency despite the existence of many complex and interacting factors in such an environment.

show abstract

“…11 A ratio of 1 indicates no effect on performance, a ratio of 10 Soot is unable to read in classfiles that include jsr instructions with no matching ret. This is not a limitation of Dava itself but we marked it as having crashed on the CB2JI obfuscation because of this.…”

Section: Impact Of Obfuscations On Performancementioning

confidence: 99%

“…As shown by recent empirical studies by Gu et al [9,10], small variations in code layout can lead to relatively large performance differences in Java (on the order of 5-10%). Thus, we can expect some performance differences between the original and obfuscated code just because the obfuscated code leads to different code layouts.…”

Section: Impact Of Obfuscations On Performancementioning

confidence: 99%

Obfuscating Java: The Most Pain for the Least Gain

Batchelder

Hendren

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Bytecode, Java's binary form, is relatively high-level and therefore susceptible to decompilation attacks. An obfuscator transforms code such that it becomes more complex and therefore harder to reverse engineer. We develop bytecode obfuscations that are complex to reverse engineer but also do not significantly degrade performance. We present three kinds of techniques that: (1) obscure intent at the operational level; (2) complicate control flow and objectoriented design (i.e. program structure); and (3) exploit the semantic gap between what is legal in source code and what is legal in bytecode. Obfuscations are applied to a benchmark suite to examine their affect on runtime performance, control flow graph complexity and decompilation. These results show that most of the obfuscations have only minor negative performance impacts and many increase complexity. In almost all cases, tested decompilers fail to produce legal source code or crash completely. Those obfuscations that are decompilable greatly reduce the readability of output source.

show abstract

Relative factors in performance analysis of Java virtual machines

Cited by 21 publications

References 30 publications

A comprehensive evaluation of object scanning techniques

A comprehensive evaluation of object scanning techniques

Phase-based adaptive recompilation in a JVM

Obfuscating Java: The Most Pain for the Least Gain

Contact Info

Product

Resources

About