(*58()9$"2#$:0/,;58(03<10/2,>=?33@">"29 #A:0*/,B58(*C2"258/052,D3*>#$,,6-*0'/ 58@F,058*,+HG?!"*0"I"252J58K0/ ,6-*0'/ 030"6N*IO40"58DP)"58QF,058SRUT6252,D<0!2T6252,V52!8("9 "W5X3,06*9E,'Y58(*03C:0'/ X3,06*9E,'Y58(*03C 1622 *'\,20/2XD3Q#$,U-0/269EU,/52,X"58QF,0'58,+ I,2/2-K58X^528-3L2T6252,_0/252/,58('4-*0'2,Y 0C#$,058Z#>58,0@=`58a02T/2*(*C/,':b(/,058c+ \",25C0d@"3,152058[#;58!*03e0/252,/58( 5805f8(""52<00"58>b(3589$3,3*"*58QF058C-02,;"(3T Y2520'58258/,03@20'Q"3+ ] D,Q"...

Instrumenting code to collect profiling information can cause substantial execution overhead. This overhead makes instrumentation difficult to perform at runtime, often preventing many known offline feedback-directed optimizations from being used in online systems. This paper presents a general framework for performing instrumentation sampling to reduce the overhead of previously expensive instrumentation. The framework is simple and effective, using code-duplication and counter-based sampling to allow switching between instrumented and non-instrumented code.

This paper describes the implementation of an online feedback-directed optimization system. The system is fully automatic; it requires no prior...

Software systems have been using “just-in-time ” compilation (JIT) techniques since the 1960s. Broadly, JIT compilation includes any translation performed dynamically, after a program has started execution. We examine the motivation behind JIT compilation and constraints imposed on JIT compilation systems, and present a classification scheme for

This paper presents a system that provides code generation at load-time and continuous program optimization at run-time. First, the architecture of the system is presented. Then, two optimization techniques are discussed that were developed specifically in the context of continuous optimization. The first of these optimizations continually adjusts the storage layouts of dynamic data structures to maximize data cache locality, while the second performs profile-driven instruction re-scheduling to increase instruction-level parallelism. These two optimizations have very di#erent cost/benefit ratios, presented in a series of benchmarks. The paper concludes with an outlook to future research directions and an enumeration of some remaining research problems. The empirical results presented in this paper make a case in favor of continuous optimization, but indicate that it needs to be applied judiciously. In many situations, the costs of dynamic optimizations outweigh their benefit, so that no break-even point is ever reached. In favorable circumstances, on the other hand, speed-ups of over 120% have been observed. It appears as if the main beneficiaries of continuous optimization are shared libraries, which at di#erent times can be optimized in the context of the currently dominant client application.

this article, we present an optimization technique that increases memory performance specifically for pointer-centric applications. Our optimization is based on determining the best internal storage layout for dynamically allocated data structures. It applies to programming languages that are fully type-safe, such as Java Parts of this work are funded by a CAREER award from the National Science Foundation (CCR-- 97014000) and by the California MICRO Program with industrial sponsor Microsoft Research (Project No. 99-039). Authors' addresses: T. Kistler, Transmeta Corporation, 3940 Freedom Circle, Santa Clara, CA 95054; M. Franz, Department of Information and Computer Science, University of California at Irvine, Irvine, CA 92697--3425.

Today’s virtual machines (VMs) dynamically optimize an application as it is executing, often employing optimizations that are specialized for the current execution profile. An online phase detector determines when an executing program is in a stable period of program execution (a phase) or is in transition. A VM using an online phase detector can apply specialized optimizations during a phase or reconsider optimization decisions between phases. Unfortunately, extant approaches to detecting phase behavior rely on either offline profiling, hardware support, or are targeted toward a particular optimization. In this work, we focus on the enabling technology of online phase detection. More specifically, we contribute (a) a novel framework for online phase detection, (b) multiple instantiations of the framework that produce novel online phase detection algorithms, (c) a novel client- and machine-independent baseline methodology for evaluating the accuracy of an online phase detector, (d) a metric to compare online detectors to this baseline, and (e) a detailed empirical evaluation, using Java applications, of the accuracy of the numerous phase detectors. 1

Virtual machines face significant performance challenges beyond those confronted by traditional static optimizers. First, portable program representations and dynamic language features, such as dynamic class loading, force the deferral of most optimizations until runtime, inducing runtime optimization overhead. Second, modular

This paper describes an empirical study of the SPECjvm98 benchmarks, using the Jalape~no virtual machine. The study employs two compilers, a nonoptimizing compiler that is initially used to compile all application methods, and an optimizing compiler that is selectively used to recompile a parameterized set of hot methods based on past profiling. We view this study as a step in examining the feasibility of adaptive optimization in this environment. The results show promise for adaptive optimization. In particular, they show that the combined time (execution and compilation) of selective opt-compilation can be less than the execution time of no opt-compilation and the combined time of full opt-compilation. The results also show that the combined time of selective opt-compilation can be competitive with static compilation (full opt-compilation not counting compilation time) for the SPECjvm98 benchmarks with input size 100. 1 Introduction One technique for increasing the efficie...

This paper provides details of the component of the Jalape~no adaptive optimization system that determines what methods to optimize. This component, called the controller , can choose from one of several optimization levels. In the current implementation, the controller uses a simple cost/benet analysis to drive adaptive compilation decisions. It has been demonstrated that even this simple analytic model can achieve reasonable performance compared to various JIT compilation scenarios in both startup and steady-state program regimes. This paper outlines several open questions in developing a more accurate controller model. We present two experiments that study the eects of how the current model predicts future execution from the past, a limited experimental evaluation of stability of the current model across applications, and describe our ongoing eorts to improve the Jalape~no controller. 1.