| Suganuma, T.; Ogasawara, T.; Takeuchi, M.; Yasue, T.; Kawahito, M.; Ishizaki, K.; Komatsu, H.; Nakatani, T. "Overview of the IBM Java just-in-time compiler", IBM Systems Journal, 2000, 39(1): 175193. |
....should by design result in better performance, but is however inherently non portable. This means that optimizations found in compilers and standard JVMs are not bene ted from: in a recent publication [2] the authors report that, in absence of denial of service attack, IBM s compiler and JVM [20] is 2 5 times faster than theirs. Developed by the same team as Ka eOS, Alta [24] is a prototype based on the Fluke hierarchical process model, and implemented on the Ka e virtual machine. The main differences with Ka eOS are that a single garbage collector is responsible for all applications, ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time compiler. ### ####### #######, 39(1):175-193, 2000.
....The Ka#eOS approach should by design result in better performance, but is however inherently non portable. This also means that optimizations found in compilers and standard JVMs are not benefited from: The authors report that, in absence of denial of service attacks, IBM s compiler and JVM [19] is 2 5 times faster than theirs. In contrast, our fully portable implementation of resource accounting in Java executes on every standard JVM and incurs only moderate overhead. NOMADS [22] is a mobile agent system, which has the ability to control resources used by agents, including protection ....
....i.e. all byte code manipulations can be done using a Java program. There are several kinds of applications where byte code engineering can be applied, such as: Optimization: To correct ine#cient byte code generated by a compiler. Such technique is used for example in Just In Time (JIT) [19] compilers that are integrated in the JVM, or can be performed o# line. It is also possible to remove unnecessary information (used only for debugging) Security: Obfuscation of applications to avoid black box testing attacks (deducing what an application do by running it several times for ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time compiler. IBM Systems Journal, 39(1):175--193, 2000. 2, 1.1.1
....include in our comparison the version of Ka#e on which Ka#eOS is based. We use a development snapshot from June 2000 for that purpose; we label this version Ka#e 2000 in our benchmarks. We also include IBM s JVM from the IBM JDK 1.1. 8, which provides one of the fastest commercial JIT compilers [73] available for JVMs that implement Java version 1.1.x. Our results are not comparable with any published SPEC JVM98 metrics, as our measurements are not compliant with all of SPEC s run rules. Complying with the run rules would have required support for the abstract windowing toolkit (AWT) which ....
Suganuma, T., Ogasawara, T., Takeuchi, M., Yasue, T., Kawahito, M., Ishizaki, K., Komatsu, H., and Nakatani, T. Overview of the IBM Java just-in-time compiler. IBM Systems Journal 39, 1 (2000), 175--193.
....mode, the per bytecode execution time is already quite high, so extra checking time has much less impact than it does in JIT3 mode. The Kaffe JIT3 compiler does not incorporate more recent and sophisticated technologies to optimize the generated code, such as those employed in IBM Jalepeno [35] and Jackal [38] to eliminate redundant object reference and array boundary checks. By applying similar techniques in J InterWeave to eliminate redundant instrumentation, we believe that the overhead could be further reduced. 5.2 Translation cost As described in Sections 3, a J InterWeave ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-In-Time Compiler. IBM Systems Journal, 39(1), 2000.
....machine and then executed. At the client, the bytecode is converted to native machine code and executed by a virtual machine. Initially, such programs were interpreted. However, to overcome the performance limitations interpretation usually imposes, these systems now employ Just InTime compilation [13, 1, 11, 7]. These virtual machines dynamically compile the bytecode stream as each new method is initially invoked. The resulting execution time is lower than for interpreted bytecodes, but execution must pause each time a method is initially invoked so that it may be compiled. Despite the use of ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175-193, 2000. 23
....code for a function is decompressed at most once the first time it is called with subsequent calls bypassing the decompressor and entering the decompressed code directly. This conceptually resembles the behavior of just in time compilers that translate interpretable code to native code [1, 22]. An alternative is to discard the decompressed code for a function when it is no longer on the call stack, since at this point we can be certain that any function called by it has returned to it already. This is the approach taken by Lucco [19] though rather than immediately discarding a ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, K. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM
....an architecture independent program representation (e.g. bytecodes for the Java language) is shipped to the execution site and interpreted by a virtual machine. However, to overcome the performance limitations interpretation usually imposes, these systems now employ just in time compilation [25, 2, 19, 11]. These new virtual machines dynamically compile the bytecode stream (on a method by method basis) into machine code before executing it. The resulting execution time is lower than for interpreted bytecodes, but execution must pause each time a method is initially invoked so that it may be ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1), 2000.
....of array regions is 24 MByte s (768 user bytes per 1K packet) Jackal s compiler generates good sequential code; sequential speed of code without access checks is at least as good as the performance of IBM s JIT version 1. 3 for Linux, which is the fastest JIT compiler system currently available [9, 27]. Generation of access checks without optimization creates a large performance penalty: up to a factor of 5.5 for the applications described below. The compiler optimization passes reduce the overhead for access checks to 9 on average for these applications. 5.1 Application Suite Our ....
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175--193, 2000. Online at http://www.research.ibm.com/journal/sj39-1.html.
....operations (at least 32 and more than 10 times of the number of call sites) but not any double precision operations, and the Rounding approach is applied for other methods. 4. 2 Environment We implemented precision region optimization on our production quality Java Just In Time (JIT) compiler [18, 11, 10, 14, 19], which is part of the IBM Developer Kit for Windows, Java 2 Technology Edition [7] Version 1.3.1. Throughout the measurements, we used the same parameters for the Java VM and SPECjvm98. The initial and maximum amounts of Java heap space were 128 MB, specified with the parameters Xms128m ....
SUGANUMA, T., OGASAWARA, T., TAKEUCHI, M., YASUE, T., KAWAHITO, M., ISHIZAKI, K., KO- MATSU, H., AND NAKATANI, T. Overview of the IBM Java Just-In-Time compiler. IBM Syst. J. 39, 1 (2000), 175--193.
No context found.
Suganuma, T.; Ogasawara, T.; Takeuchi, M.; Yasue, T.; Kawahito, M.; Ishizaki, K.; Komatsu, H.; Nakatani, T. "Overview of the IBM Java just-in-time compiler", IBM Systems Journal, 2000, 39(1): 175193.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java just-in-time compiler. IBM Systems Journals, Java Performance Issue, 39(1), February 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175--193, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175--193, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175-193, 2000.
No context found.
Suganuma, T., Ogasawara, T., Takeuchi, M., Yasue, T., Kawahito, M., Ishizaki, K., Komatsu, H., , Nakatani, T.: Overview of the IBM Java just-in-time compiler. IBM Systems Journals 39 (2000) 175--193
No context found.
Suganuma, T., Ogasawara, T., Takeuchi, M., Yasue, T., Kawahito, M., Ishizaki, K., Komatsu, H., , Nakatani, T.: Overview of the IBM Java just-in-time compiler. IBM Systems Journals 39 (2000) 175--193
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1), 2000.
No context found.
Suganuma, T., Ogasawara, T., Takeuchi, M., Yasue, T., Kawahito, M., Ishizaki, K., Komatsu, H., , Nakatani, T.: Overview of the IBM Java just-in-time compiler. IBM Systems Journals 39 (2000) 175--193
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java just-in-time compiler. IBM Systems Journal, 39(1):175--193, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani, "Overview of the IBM Java Just-in-Time Compiler," IBM Systems Journal, vol. 39, no. 1, pp. 175--193, Feb. 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175--193, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatso, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175-- 193, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatsu and T. Nakatani, "Overview of the IBM JAva Just-in-Time Compiler", IBM Systems Journal, vol. 39, pp. 175-193, 2000.
No context found.
Suganuma et al. Overview of the IBM Java Just-In-time Compiler, IBM Systems Journal, Vol. 39, No. 1, 2000.
No context found.
T. Suganuma, T. Ogasawara, M. Takeuchi, T. Yasue, M. Kawahito, K. Ishizaki, H. Komatso, and T. Nakatani. Overview of the IBM Java Just-in-Time Compiler. IBM Systems Journal, 39(1):175--193, 2000. http://www.research.ibm.com/journal/sj/391/suganuma.html.
First 50 documents
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC