Online feedback-directed optimization of Java

Arnold, Matthew; Hind, Michael; Ryder, Barbara G.

doi:10.1145/582431.582432

Cited by 25 publications

(41 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Examples of adaptive optimization systems include HotSpot virtual machine [64] from Sun (now Oracle), Jikes RVM [65] from IBM, and Open Runtime Platform (ORP) [66] from Intel Corporation. To identify hot-spots, researchers have proposed to use online hardware profiling mechanisms such as counters and samplings [67][68][69][70][71], or to use program instrumentation [72][73][74][75][76][77][78], combined instrumentation and sampling [79][80][81], or coupled offline and online profiling [82]. To further improve adaptive optimization, a number of techniques have been developed; for example, recompilation [83], deferred and partial compilation [84][85][86], and dynamic deoptimization [87].…”

Section: B Jit Compilationmentioning

confidence: 99%

Overview of Real-Time Java Computing

Sun¹,

Zhang²

2013

Journal of Computing Science and Engineering

View full text Add to dashboard Cite

This paper presents a complete survey of recent techniques that are applied in the field of real-time Java computing. It focuses on the issues that are especially important for hard real-time applications, which include time predictable garbage collection, worst-case execution time analysis of Java programs, real-time Java threads scheduling and compiler techniques designed for real-time purpose. It also evaluates experimental frameworks that can be used for researching real-time Java. This overview is expected to help researchers understand the state-of-the-art and advance the research in real-time Java computing.

show abstract

Section: B Jit Compilationmentioning

confidence: 99%

Overview of Real-Time Java Computing

Sun¹,

Zhang²

2013

Journal of Computing Science and Engineering

View full text Add to dashboard Cite

show abstract

“…A prototype compiler is presented in [3], where loop transformation and parallelization techniques are used to achieve high performance on numerical Java codes. Jikes [2] compiler uses runtime feedback to direct its adaptive optimization. Dynamic optimization is presented in [21] to measure the cost of thread creation and parallelize code insides a Jikes RVM at run time.…”

Section: Related Workmentioning

confidence: 99%

A Cost-Aware Parallel Workload Allocation Approach Based on Machine Learning Techniques

Long

Fursin

Franke

2007

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Parallelism is one of the main sources for performance improvement in modern computing environment, but the efficient exploitation of the available parallelism depends on a number of parameters. Determining the optimum number of threads for a given data parallel loop, for example, is a difficult problem and dependent on the specific parallel platform. This paper presents a learning-based approach to parallel workload allocation in a costaware manner. This approach uses static program features to classify programs, before deciding the best workload allocation scheme based on its prior experience with similar programs. Experimental results on 12 Java benchmarks (76 test cases with different workloads in total) show that it can efficiently allocate the parallel workload among Java threads and achieve an efficiency of 86% on average.

show abstract

“…Most existing profiling techniques are meant for off-line program analysis. However, with the advent of dynamic compilation and runtime optimizations, the use of profile data generated for runtime use has increased [19,16,13,3,2,5,4]. Recently [7] has shown that runtime flow-specific information can be used to improve code performance significantly.…”

Section: Background and Motivationmentioning

confidence: 99%

“…Of these a small number of paths (≈ 16 for K=50 and ≈ 14 for K=100) account for a fairly large percentage (≈ 61% for K=50 and ≈ 59% for K=100) of the total iterations in these regions at runtime. 4 These paths also have the property that the code continuously stays in these paths for at least 50/100 iterations on average without shifting to the other possible paths in the region. Thus it makes sense to perform path-specific runtime optimizations on these paths since (i) these paths constitute a fair fraction of the executed code and (ii) the path-specific optimizations will hold true for that period, allowing them to be profitable.…”

Section: Statistics From Tfpmentioning

confidence: 99%

TFP: Time-Sensitive, Flow-Specific Profiling at Runtime

Nandy

Gao

Ferrante

2004

Languages and Compilers for Parallel Computing

View full text Add to dashboard Cite

Abstract. Program profiling can help performance prediction and compiler optimization. This paper describes the initial work behind TFP, a new profiling strategy that can gather and verify a range of flow-specific information at runtime. While TFP can collect more refined information than block, edge or path profiling, it is only 5.75% slower than a very fast runtime path-profiling technique. Statistics collected using TFP over the SPEC2000 benchmarks reveal possibilities for further flow-specific runtime optimizations. We also show how TFP can improve the overall performance of a real application.

show abstract

Online feedback-directed optimization of Java

Cited by 25 publications

References 0 publications

Overview of Real-Time Java Computing

Overview of Real-Time Java Computing

A Cost-Aware Parallel Workload Allocation Approach Based on Machine Learning Techniques

TFP: Time-Sensitive, Flow-Specific Profiling at Runtime

Contact Info

Product

Resources

About