A study of the Allocation Behavior of the SPECjvm98 Java Benchmarks

Dieckmann, Sylvia; Hölzle, Urs

doi:10.1007/3-540-48743-3_5

Cited by 116 publications

(120 citation statements)

References 18 publications

Supporting

Mentioning

109

Contrasting

Unclassified

Order By: Relevance

“…It is worth noticing that typical objects in object-oriented languages such as Java and C# only have an average object size of less than 100 bytes [7,21]. However, the average size of the compiled methods in each application range from 300 bytes to over 1100 bytes.…”

Section: Rq1: Access Behaviormentioning

confidence: 99%

An Empirical Study of the Code Pitching Mechanism in the .NET Framework.

Anthony¹,

Srisa-an²,

Leung³

2006

JOT

View full text Add to dashboard Cite

The .NET Compact Framework is designed to be a high-performance virtual machine for mobile and embedded devices that operate on Windows CE (version 4.1 and later). It achieves fast execution time by compiling methods dynamically instead of using interpretation. Once compiled, these methods are stored in a portion of the heap called the code cache and can be quickly reused to satisfy future method calls. While the code cache provides a high-level of reusability, it can also use a large amount of memory. As a result, the Compact Framework provides a "code pitching" mechanism that can be used to discard the previously compiled methods as needed. In this paper, we study the effect of code pitching on the overall performance and memory utilization of .NET applications. We conduct our experiments using Microsoft's Shared-Source Common Language Infrastructure (SSCLI). We profile the access behavior of the compiled methods. We also experiment with various code cache configurations to perform pitching. We find that programs can operate efficiently with a small code cache without incurring substantial recompilation and execution overheads.

show abstract

Section: Rq1: Access Behaviormentioning

confidence: 99%

An Empirical Study of the Code Pitching Mechanism in the .NET Framework.

Anthony¹,

Srisa-an²,

Leung³

2006

JOT

View full text Add to dashboard Cite

show abstract

“…The SPEC JVM98 suite is perhaps more commonly used to measure the speed and effectiveness of Java compilers and virtual machines [5,6,23,27,31]. Other views include those discussing lower-level architectural issues relating to the SPEC programs [20,24], allocation and heap behaviour [12], as well as measuring the performance of various Java microarchitectures [22] .…”

Section: The Benchmark Suitesmentioning

confidence: 99%

A method‐level comparison of the Java Grande and SPEC JVM98 benchmark suites

Gregg

Power

Waldron

2005

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYIn this paper we seek to provide a foundation for the study of the level of use of objectoriented techniques in Java programs in general, and scientific applications in particular. In particular we investigate the profiles of Java programs from a number of perspectives, including the use of class library methods, the size of methods called, the mode of invoke instruction used and the polymorphicity of call sites. We also present a categorisation of the nature of small-sized methods used in Java programs. We compare the Java Grande and SPEC JVM98 benchmark suites, and note a significant difference in the nature and composition of these suites, with the programs from the Java Grande suite demonstrating a less object-oriented approach.

show abstract

“…While the extensive study of SPECjvm98 [26] benchmarks by [4] yielded many insights into the allocation and garbage collection behavior of Java programs, SPECjvm98 does not represent distributed application and server workloads. In [32], a study was performed to investigate whether the CLI implementations can be "useful for high-performance computing."…”

Section: Related Workmentioning

confidence: 99%

Remote Objects: The Next Garbage Collection Challenge.

Srisa-an¹,

Oey²

2005

JOT

View full text Add to dashboard Cite

In this paper, we present an analysis of remote objects in Microsoft's Shared Source Common Language Infrastructure. The contribution of our work is threefold. First, we analyze the behavior of remote objects. We find that the basic behavior of these objects significantly differ from the ones of local objects, which have been thoroughly studied in the past. We also study the behavior of remote objects with respect to different activation modes (i.e. single-call, singleton, and client-activated). Second, based on those behavioral differences, we study the impacts of managing remote objects with a generational garbage collector that is designed essentially to manage local objects. We find that the garbage collection efficiency degrades significantly when the heap is interspersed with both local and remote objects. Third, we suggest various optimization techniques to improve the garbage collection efficiency in distributed objects environments.

show abstract

A study of the Allocation Behavior of the SPECjvm98 Java Benchmarks

Cited by 116 publications

References 18 publications

An Empirical Study of the Code Pitching Mechanism in the .NET Framework.

An Empirical Study of the Code Pitching Mechanism in the .NET Framework.

A method‐level comparison of the Java Grande and SPEC JVM98 benchmark suites

Remote Objects: The Next Garbage Collection Challenge.

Contact Info

Product

Resources

About