Cloud developers have to build applications that are resilient to failures and interruptions. We advocate for a fault-tolerant programming model for the cloud based on actors, retry orchestration, and tail calls. This model builds upon persistent data stores and message queues readily available on the cloud. Retry orchestration not only guarantees that (1) failed actor invocations will be retried but also that (2) completed invocations are never repeated and (3) it preserves a strict happen-before relationship across failures within call stacks. Tail calls can break complex tasks into simple steps to minimize re-execution during recovery. We review key application patterns and failure scenarios. We formalize a process calculus to precisely capture the mechanisms of fault tolerance in this model. We briefly describe our implementation. Using an application inspired by a typical enterprise scenario, we validate the functional correctness of our implementation and assess the impact of fault preparedness and recovery on performance.

Cloud-based applications are ubiquitous and essential. We expect them to be simultaneously scalable and available and simple to build and deploy. Virtual Programming Environments are what make these applications possible. Virtual Programming Environments are themselves complex distributed systems, built using the entire spectrum of System and Runtime Virtualization technology that is the subject of VEE. In the first part of the talk, I will focus on the purest form of Serverless Computing: Functions as a Service (FaaS) as embodied in Virtual Programming Environments such as AWS Lambda or Apache OpenWhisk. I will describe the programming abstractions they provide to the developer and how these abstractions are realized using virtualization technology. In the second part of the talk, I will outline the research challenges in moving beyond FaaS to build Virtual Programming Environments that can productively support building complex stateful applications on the cloud.

Cloud computing has made the resources needed to execute large-scale in-memory distributed computations widely available. Specialized programming models, e.g., MapReduce, have emerged to offer transparent fault tolerance and fault recovery for specific computational patterns, but they sacrifice generality. In contrast, the Resilient X10 programming language adds failure containment and failure awareness to a general purpose, distributed programming language. A Resilient X10 application spans over a number of places. Its formal semantics precisely specify how it continues executing after a place failure. Thanks to failure awareness, the X10 programmer can in principle build redundancy into an application to recover from failures. In practice, however, correctness is elusive, as redundancy and recovery are often complex programming tasks. This article further develops Resilient X10 to shift the focus from failure awareness to failure recovery, from both a theoretical and a practical standpoint. We rigorously define the distinction between recoverable and catastrophic failures. We revisit the happens-before invariance principle and its implementation. We shift most of the burden of redundancy and recovery from the programmer to the runtime system and standard library. We make it easy to protect critical data from failure using resilient stores and harness elasticity—dynamic place creation—to persist not just the data but also its spatial distribution. We demonstrate the flexibility and practical usefulness of Resilient X10 by building several representative high-performance in-memory parallel application kernels and frameworks. These codes are 10× to 25× larger than previous Resilient X10 benchmarks. For each application kernel, the average runtime overhead of resiliency is less than 7%. By comparing application kernels written in the Resilient X10 and Spark programming models, we demonstrate that Resilient X10’s more general programming model can enable significantly better application performance for resilient in-memory distributed computations.

Swift is a modern multi-paradigm programming language with an extensive developer community and open source ecosystem. Swift 3's memory management strategy is based on Automatic Reference Counting (ARC) augmented with unsafe APIs for manually-managed memory. We have seen ARC consume as much as 80% of program execution time. A significant portion of ARC's direct performance cost can be attributed to its use of atomic machine instructions to protect reference count updates from data races. Consequently, we have designed and implemented dynamic atomicity, an optimization which safely replaces atomic reference-counting operations with nonatomic ones where feasible. The optimization introduces a store barrier to detect possibly intra-thread references, compiler-generated recursive reference-tracers to find all affected objects, and a bit of state in each reference count to encode its atomicity requirements.

Scale-out programs run on multiple processes in a cluster. In scale-out systems, processes can fail. Computations using traditional libraries such as MPI fail when any component process fails. The advent of Map Reduce, Resilient Data Sets and MillWheel has shown dramatic improvements in productivity are possible when a high-level programming framework handles scale-out and resilience automatically.

X10 is a high-performance, high-productivity programming language aimed at large-scale distributed and shared-memory parallel applications. It is based on the Asynchronous Partitioned Global Address Space (APGAS) programming model, supporting the same fine-grained concurrency mechanisms within and across shared-memory nodes.

This paper addresses the problem of efficiently supporting parallelism within a managed runtime. A popular approach for exploiting software parallelism on parallel hardware is task parallelism, where the programmer explicitly identifies potential parallelism and the runtime then schedules the work. Work-stealing is a promising scheduling strategy that a runtime may use to keep otherwise idle hardware busy while relieving overloaded hardware of its burden. However, work-stealing comes with substantial overheads. Recent work identified sequential overheads of work-stealing, those that occur even when no stealing takes place, as a significant source of overhead. That work was able to reduce sequential overheads to just 15%.

Work-stealing is a promising approach for effectively exploiting software parallelism on parallel hardware. A programmer who uses work-stealing explicitly identifies potential parallelism and the runtime then schedules work, keeping otherwise idle hardware busy while relieving overloaded hardware of its burden. Prior work has demonstrated that work-stealing is very effective in practice. However, work-stealing comes with a substantial overhead: as much as 2x to 12x slowdown over orthodox sequential code.

SUMMARY Programming for large‐scale, multicore‐based architectures requires adequate tools that offer ease of programming and do not hinder application performance. StarSs is a family of parallel programming models based on automatic function‐level parallelism that targets productivity. StarSs deploys a data‐flow model: it analyzes dependencies between tasks and manages their execution, exploiting their concurrency as much as possible. This paper introduces Cluster Superscalar (ClusterSs), a new StarSs member designed to execute on clusters of SMPs (Symmetric Multiprocessors). ClusterSs tasks are asynchronously created and assigned to the available resources with the support of the IBM APGAS runtime, which provides an efficient and portable communication layer based on one‐sided communication. We present the design of ClusterSs on top of APGAS, as well as the programming model and execution runtime for Java applications. Finally, we evaluate the productivity of ClusterSs, both in terms of programmability and performance and compare it to that of the IBM X10 language. Copyright © 2012 John Wiley & Sons, Ltd.

Programming for large-scale, multicore-based architectures requires adequate tools that offer ease of programming while not hindering application performance. StarSs is a family of parallel programming models based on automatic function level parallelism that targets productivity. StarSs deploys a data-flow model: it analyses dependencies between tasks and manages their execution, exploiting their concurrency as much as possible. We introduce Cluster Superscalar (ClusterSs), a new StarSs member designed to execute on clusters of SMPs. ClusterSs tasks are asynchronously created and assigned to the available resources with the support of the IBM APGAS runtime, which provides an efficient and portable communication layer based on one-sided communication.This short paper gives an overview of the ClusterSs design on top of APGAS, as well as the conclusions of a productivity study; in this study, ClusterSs was compared to the IBM X10 language, both in terms of programmability and performance. A technical report is available with the details.

X10 is a new object-oriented PGAS (Partitioned Global Address Space) programming language with support for distributed asynchronous dynamic parallelism that goes beyond past SPMD message-passing models such as MPI and SPMD PGAS models such as UPC and Co-Array Fortran. The concurrency constructs in X10 make it possible to express complex computation and communication structures with higher productivity than other distributed-memory programming models. However, this productivity often comes at the cost of high performance overhead when the language is used in its full generality. This paper introduces high-level compiler optimizations and transformations to reduce communication and synchronization overheads in distributed-memory implementations of X10 programs. Specifically, we focus on locality optimizations such as scalar replacement and task localization, combined with supporting transformations such as loop distribution, scalar expansion, loop tiling, and loop splitting. We have completed a prototype implementation of these high-level optimizations, and performed a performance evaluation that shows significant improvements in performance, scalability, communication volume and number of tasks. We evaluated the communication optimizations on three platforms: a 128-node Blue Gene/P cluster, a 32-node Nehalem cluster, and a 16-node Power7 cluster. On the Blue Gene/P cluster, we observed a maximum performance improvement of 31.46x relative to the unoptimized case (for the MolDyn benchmark). On the Nehalem cluster, we observed a maximum performance improvement of 3.01x (for the NQueens benchmark) and on the Power7 cluster, we observed a maximum performance improvement of 2.73x (for the MolDyn benchmark). In addition, there was no case in which the optimized code was slower than the unoptimized case. We also believe that the optimizations presented in this paper will be necessary for any high-productivity PGAS language based on modern object-oriented principles, that is designed for execution on future Extreme Scale systems that place a high premium on locality improvement for performance and energy efficiency.

X10 is an emerging Partitioned Global Address Space (PGAS) language intended to increase significantly the productivity of developing scalable HPC applications. The language has now matured to a point where it is meaningful to consider writing large scale scientific application codes in X10. This paper reports our experiences writing three codes from the chemistry/material science domain: Fast Multipole Method (FMM), Particle Mesh Ewald (PME) and Hartree-Fock (HF), entirely in X10. Performance results are presented for up to 256 places on a Blue Gene/P system. During the course of this work our experiences have been shared with the X10 development team, so that application requirements could inform language design discussions as the language capabilities influenced algorithm design. This resulted in improvements in the language implementation and standard class libraries, including the design of the array API and support for complex math. Data constructs in X10 such as \emph{places} and \emph{distributed arrays}, and parallel constructs such as \emph{finish} and \emph{async}, simplify implementation of the applications in comparison with MPI. However, current implementation limitations in X10 2.1.2 make it difficult to achieve scalable performance using the most natural expressions of the algorithms. The most serious limitation is the use of point-to-point communication patterns, rather than collectives, to implement parallel constructs and array operations. This issue will be addressed in future releases of X10.

On shared-memory systems, Cilk-style work-stealing has been used to effectively parallelize irregular task-graph based applications such as Unbalanced Tree Search (UTS). There are two main difficulties in extending this approach to distributed memory. In the shared memory approach, thieves (nodes without work) constantly attempt to asynchronously steal work from randomly chosen victims until they find work. In distributed memory, thieves cannot autonomously steal work from a victim without disrupting its execution. When work is sparse, this results in performance degradation. In essence, a direct extension of traditional work-stealing to distributed memory violates the work-first principle underlying work-stealing. Further, thieves spend useless CPU cycles attacking victims that have no work, resulting in system inefficiencies in multi-programmed contexts. Second, it is non-trivial to detect active distributed termination (detect that programs at all nodes are looking for work, hence there is no work). This problem is well-studied and requires careful design for good performance. Unfortunately, in most existing languages/frameworks, application developers are forced to implement their own distributed termination detection.

The power of high-level languages lies in their abstraction over hardware and software complexity, leading to greater security, better reliability, and lower development costs. However, opaque abstractions are often show-stoppers for systems programmers, forcing them to either break the abstraction, or more often, simply give up and use a different language. This paper addresses the challenge of opening up a high-level language to allow practical low-level programming without forsaking integrity or performance.

Real-time Garbage Collection (RTGC) has recently advanced to the point where it is being used in production for financial trading, military command-and-control, and telecommunications. However, among potential users of RTGC, there is enormous diversity in both application requirements and deployment environments.

The emergence of standards for programming real-time systems in Java has encouraged many developers to consider its use for systems previously only built using C, Ada, or assembly language. However, the RTSJ standard in isolation leaves many important problems unaddressed, and suffers from some serious problems in usability and safety.

If the operating system could be specialized for every application, many applications would run faster. For example, Java virtual machines (JVMs) provide their own threading model and memory protection, so general-purpose operating system implementations of these abstractions are redundant. However, traditional means of transforming existing systems into specialized systems are difficult to adopt because they require replacing the entire operating system. This paper describes Libra, an execution environment specialized for IBM's J9 JVM. Libra does not replace the entire operating system. Instead, Libra and J9 form a single statically-linked image that runs in a hypervisor partition. Libra provides the services necessary to achieve good performance for the Java workloads of interest but relies on an instance of Linux in another hypervisor partition to provide a networking stack, a filesystem, and other services. The expense of remote calls is offset by the fact that Libra's services can be customized for a particular workload; for example, on the Nutch search engine, we show that two simple customizations improve application throughput by a factor of 2.7.

Poor instruction cache locality can degrade performance on modern architectures. For example, our simulation results show that eliminating all instruction cache misses improves performance by as much as 16% for a modestly sized instruction cache. In this paper, we show how to take advantage of dynamic code generationin a Java Virtual Machine (VM) to improve instruction locality at run-time. We develop a dynamic code reordering (DCR) system; alow overhead, online approach for improving instruction locality. DCR has three optimizations: (1) Interprocedural method separation; (2) Intraprocedural code splitting; and (3) Code padding. DCR uses the dynamic call graph and an edge profile that most VMs already collect to separate hot/cold methods and hot/cold code within a method. It also puts padding between methods to minimize conflict misses between frequent caller/callee pairs. It incrementally performs these optimizations only when the VM is optimizing a method at a higher level. We implement DCR in Jikes RVM and show its overhead is negligible. Extensive simulation and run-time experiments show that a simple code space improves average performance on a Pentium 4 by around 6% on SPEC and DaCapo Java benchmarks. These programs however have very small instruction cache footprints that limit opportunities for DCR to improve performance. Consequently, DCR optimizations on average show little effect, sometimes degrading performance and occasionally improving performance by up to 5%. Our work shows that the VM has the potential to dynamically improve instruction locality incrementally by simply piggybacking on hotspot recompilation.

While real-time garbage collection has achieved worst-case latencies on the order of a millisecond, this technology is approaching its practical limits. For tasks requiring extremely low latency, and especially periodic tasks with frequencies above 1 KHz, Java programmers must currently resort to the NoHeapRealtimeThread construct of the Real-Time Specification for Java. This technique requires expensive run-time checks, can result in unpredictable low-level exceptions, and inhibits communication with the rest of the garbage-collected application. We present Eventrons, a programming construct that can arbitrarily preempt the garbage collector, yet guarantees safety and allows its data to be visible to the garbage-collected heap. Eventrons are a strict subset of Java, and require no run-time memory access checks. Safety is enforced using a data-sensitive analysis and simple run-time support with extremely low overhead. We have implemented Eventrons in IBM's J9 Java virtual machine, and present experimental results in which we ran Eventrons at frequencies up to 22 KHz (a 45 μs period). Across 10 million periods, 99.997% of the executions ran within 10 μss of their deadline, compared to 99.999% of the executions of the equivalent program written in C.

Due to the high dynamic frequency of virtual method calls in typical object-oriented programs, feedback-directed devirtualization and inlining is one of the most important optimizations performed by high-performance virtual machines. A critical input to effective feedback-directed inlining is an accurate dynamic call graph. In a virtual machine, the dynamic call graph is computed online during program execution. Therefore, to maximize overall system performance, the profiling mechanism must strike a balance between profile accuracy, the speed at which the profile becomes available to the optimizer, and profiling overhead. This paper introduces a new low-overhead sampling-based technique that rapidly converges on a high-accuracy dynamic call graph. We have implemented the technique in two high-performance virtual machines: Jikes RVM and J9. We empirically assess our profiling technique by reporting on the accuracy of the dynamic call graphs it computes and by demonstrating that increasing the accuracy of the dynamic call graph results in more effective feedback-directed inlining.

Real-time systems have reached a level of complexity beyond the scaling capability of the low-level or restricted languages traditionally used for real-time programming.While Metronome garbage collection has made it practical to use Java to implement real-time systems, many challenges remain for the construction of complex real-time systems, some specific to the use of Java and others simply due to the change in scale of such systems.The goal of our current research is the creation of a comprehensive Java-based programming environment and methodology for the creation of complex real-time systems. Our goals include construction of a provably correct real-time garbage collector capable of providing worst case latencies of 100 μs, capable of scaling from sensor nodes up to large multiprocessors; specialized programming constructs that retain the safety and simplicity of Java, and yet provide sub-microsecond latencies; the extension of Java's "write once, run anywhere" principle from functional correctness to timing behavior; on-line analysis and visualization that aids in the understanding of complex behaviors; and a principled probabilistic analysis methodology for bounding the behavior of the resulting systems.While much remains to be done, this paper describes the progress we have made towards these goals.

Real-time garbage collection has been shown to be feasible, but for programs with high allocation rates, the utilization achievable is not sufficient for some systems.Since a high allocation rate is often correlated with a more high-level, abstract programming style, the ability to provide good real-time performance for such programs will help continue to raise the level of abstraction at which real-time systems can be programmed.We have developed techniques that allow generational collection to be used despite the problems caused by variance in program behavior over the short time scales in which a nursery can be collected. Syncopation allows such behavior to be detected by the scheduler in time for allocation to by-pass the nursery and allow real-time bounds to be met.We have provided an analysis of the costs of both generational and non-generational techniques, which allow the trade-offs to be evaluated quantitatively. We have also provided measurements of application behavior which show that while syncopation is necessary, the need for it is rare enough that generational collection can provide major improvements in real-time utilization. An additional technique, arraylet pre-tenuring, often significantly improves generational behavior.

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 program representations preclude many forms of whole-program interprocedural optimization. Third, virtual machines incur additional costs for runtime services such as security guarantees and automatic memory management. To address these challenges, vendors have invested considerable resources into adaptive optimization systems in production virtual machines. Today, mainstream virtual machine implementations include substantial infrastructure for online monitoring and profiling, runtime compilation, and feedback-directed optimization. As a result, adaptive optimization has begun to mature as a widespread production-level technology. This paper surveys the evolution and current state of adaptive optimization technology in virtual machines.

Security concerns on embedded devices like cellular phones make Java an extremely attractive technology for providing third-party and user-downloadable functionality. However, garbage collectors have typically required several times the maximum live data set size (which is the minimum possible heap size) in order to run well. In addition, the size of the virtual machine (ROM) image and the size of the collector's data structures (metadata) have not been a concern for server- or workstation-oriented collectors.We have implemented two different collectors specifically designed to operate well on small embedded devices. We have also developed a number of algorithmic improvements and compression techniques that allow us to eliminate almost all of the per-object overhead that the virtual machine and the garbage collector require. We describe these optimizations and present measurements of the Java embedded benchmarks (EEMBC) of our implementations on both an IA32 laptop and an ARM-based PDA.For applications with low to moderate allocation rates, our optimized collector running on the ARM is able to achieve 85% of peak performance with only 1.05 to 1.3 times the absolute minimum heap size. For applications with high allocation rates, the collector achieves 85% of peak performance with 1.75 to 2.5 times the minimum heap size. The collector code takes up 40 KB of ROM, and collector metadata overhead has been almost completely eliminated, consuming only 0.4% of the heap.

As current trends in software development move toward more complex object-oriented programming, inlining has become a vital optimization that provides substantial performance improvements to C++ and Java programs. Yet, the aggressiveness of the inlining algorithm must be carefully monitored to effectively balance performance and code size. The state-of-the-art is to use profile information (associated with call edges) to guide inlining decisions. In the presence of virtual method calls, profile information for one call edge may not be sufficient for making effectual inlining decisions. Therefore, we explore the use of profiling data with additional levels of context sensitivity. In addition to exploring fixed levels of context sensitivity, we explore several adaptive schemes that attempt to find the ideal degree of context sensitivity for each call site. Our techniques are evaluated on the basis of runtime performance, code size and dynamic compilation time. On average, we found that with minimal impact on performance (+/-1%) context sensitivity can enable 10% reductions in compiled code space and compile time. Performance on individual programs varied from -4.2% to 5.3% while reductions in compile time and code space of up to 33.0% and 56.7% respectively were obtained.

While Java provides many software engineering benefits, it lacks a coherent module system and instead provides only packages (which are primarily a name space mechanism) and classloaders (which are very low-level). As a result, large Java applications suffer from unexpected interactions between independent components, require complex CLASSPATH definitions, and are often extremely complex to install and maintain. We have implemented a module system for Java called MJ that is implemented with class loaders, but provides a much higher-level interface. High-level properties can be specified in a module definition and are enforced by the module system as new modules are loaded. To experimentally validate the ability of MJ to properly handle the complex module inter-relationships found in large Java server systems, we replaced the classloader mechanisms of Apache Tomcat 4.1.18 [27] with 30 MJ modules. The modified Tomcat is functionally identical to the original, but requires no CLASSPATH definitions, and will operate correctly even if user code loads a different version of a module used by Tomcat, such as the Xerces XML parser [31]. Furthermore, by making a small change to the Java core libraries enabled by MJ, we obtained a 30% performance improvement in a servlet microbenchmark.

Dataflow analyses can have mutually beneficial interactions. Previous efforts to exploit these interactions have either (1) iteratively performed each individual analysis until no further improvements are discovered or (2) developed "super-analyses" that manually combine conceptually separate analyses. We have devised a new approach that allows analyses to be defined independently while still enabling them to be combined automatically and profitably. Our approach avoids the loss of precision associated with iterating individual analyses and the implementation difficulties of manually writing a super-analysis. The key to our approach is a novel method of implicit communication between the individual components of a super-analysis based on graph transformations. In this paper, we precisely define our approach; we demonstrate that it is sound and it terminates; finally we give experimental results showing that in practice (1) our framework produces results at least as precise as iterating the individual analyses while compiling at least 5 times faster, and (2) our framework achieves the same precision as a manually written super-analysis while incurring a compile-time overhead of less than 20%.

Single superclass inheritance enables simple and efficient table-driven virtual method dispathc. However, virtual method table dispatch does not handle multiple inheritance and interfaces. This complication has led to a widespread misimpression that interface method dispatch is inherently inefficient. This paper argues that with proper implementation techniques, Java interfaces need not be a source of significant performance degradation.

In this paper, we report on our experiences with guaranteeing GC-pointer safety when using unsafe low-level language extensions to implement a JVM in Java. We give an overview of the original unsafe language extensions that were defined for use by Jalapeño implementers, and introduce sanitized replacements that capture common idioms while also guaranteeing GC-pointer safety. We also outline some simple static and dynamic checks for correct usage of low-level operations, and examine how code containing low-level operations can be optimized correctly and effectively.

Abstract The execution model for mobile, dynamically‐linked, object‐oriented programs has evolved from fast interpretation to a mix of interpreted and dynamically compiled execution. The primary motivation for dynamic compilation is that compiled code executes significantly faster than interpreted code. However, dynamic compilation, which is performed while the application is running, introduces execution delay. In this paper we present two dynamic compilation techniques that enable high performance execution while reducing the effect of this compilation overhead. These techniques can be classified as (1) decreasing the amount of compilation performed, and (2) overlapping compilation with execution. We first present and evaluate lazy compilation , an approach used in most dynamic compilation systems in which individual methods are compiled on‐demand upon their first invocation. This is in contrast to eager compilation , in which all methods in a class are compiled when a new class is loaded. In this work, we describe our experience with eager compilation, as well as the implementation and transition to lazy compilation. We empirically detail the effectiveness of this decision. Our experimental results using the SpecJVM Java benchmarks and the Jalapeño JVM show that, compared to eager compilation, lazy compilation results in 57% fewer methods being compiled and reductions in total time of 14 to 26%. Total time in this context is compilation plus execution time. Next, we present profile‐driven, background compilation, a technique that augments lazy compilation by using idle cycles in multiprocessor systems to overlap compilation with application execution. With this approach, compilation occurs on a thread separate from that of application threads so as to reduce intermittent, and possibly substantial, delay in execution. Profile information is used to prioritize methods as candidates for background compilation. Methods are compiled according to this priority scheme so that performance‐critical methods are invoked using optimized code as soon as possible. Our results indicate that background compilation can achieve the performance of off‐line compiled applications and masks almost all compilation overhead. We show significant reductions in total time of 14 to 71% over lazy compilation. Copyright © 2001 John Wiley & Sons, Ltd.

A large number of call graph construction algorithms for object-oriented and functional languages have been proposed, each embodying different tradeoffs between analysis cost and call graph precision. In this article we present a unifying framework for understanding call graph construction algorithms and an empirical comparison of a representative set of algorithms. We first present a general parameterized algorithm that encompasses many well-known and novel call graph construction algorithms. We have implemented this general algorithm in the Vortex compiler infrastructure, a mature, multilanguage, optimizing compiler. The Vortex implementation provides a "level playing field" for meaningful cross-algorithm performance comparisons. The costs and benefits of a number of call graph construction algorithms are empirically assessed by applying their Vortex implementation to a suite of sizeable (5,000 to 50,000 lines of code) Cecil and Java programs. For many of these applications, interprocedural analysis enabled substantial speed-ups over an already highly optimized baseline. Furthermore, a significant fraction of these speed-ups can be obtained through the use of a scalable, near-linear time call graph construction algorithm.

Future high-performance virtual machines will improve performance through sophisticated online feedback-directed optimizations. this paper presents the architecture of the Jalapeño Adaptive Optimization System, a system to support leading-edge virtual machine technology and enable ongoing research on online feedback-directed optimizations. We describe the extensible system architecture, based on a federation of threads with asynchronous communication. We present an implementation of the general architecture that supports adaptive multi-level optimization based purely on statistical sampling. We empirically demonstrate that this profiling technique has low overhead and can improve startup and steady-state performance, even without the presence of online feedback-directed optimizations. The paper also describes and evaluates an online feedback-directed inlining optimization based on statistical edge sampling. The system is written completely in Java, applying the described techniques not only to application code and standard libraries, but also to the virtual machine itself.

The Factored Control Flow Graph, FCFG, is a novel representation of a program's intraprocedural control flow, which is designed to efficiently support the analysis of programs written in languages, such as Java, that have frequently occurring operations whose execution may result in exceptional control flow. The FCFG is more compact than traditional CFG representations for exceptional control flow, yet there is no loss of precision in using the FCFG. In this paper, we introduce the FCFG representation and outline how standard forward and backward data flow analysis algorithms can be adapted to work on this representation. We also present empirical measurements of FCFG sizes for a large number of methods obtained from a variety of Java programs, and compare these sizes with those of a traditional CFG representation.

Previous algorithms for interprocedural control flow analysis of higher-order and/or object-oriented languages have been described that perform propagation or constraint satisfaction and take O(N3) time (such as Shivers's O-CFA and Heintze's set-based analysis), or unification and take O(Nα (N,N) time (such as Steensgaard's pointer analysis), or optimistic reachability analysis and take O(N) time (such as Bacon and Sweeney's Rapid Type Analysis). We describe a general parameterized analysis framework that integrates propagation-based and unification-based analysis primitives and optimistic reachability analysis, whose instances mimic these existing algorithms as well as several new algorithms taking O(N), O(Nα(N,N)), O(N2), and O(N2α(N,N)) time; our O(N) and O(Nα(N,N)) algorithms produce more precise results than the previous algorithms with these complexities. We implemented our algorithm framework in the Vortex optimizing compiler, and we measured the cost and benefit of these interprocedural analysis algorithms in practice on a collection of substantial Cecil and Java programs.

Interprocedural analyses enable optimizing compilers to more precisely model the effects of non-inlined procedure calls, potentially resulting in substantial increases in application performance. Applying interprocedural analysis to programs written in object-oriented or functional languages is complicated by the difficulty of constructing an accurate program call graph. This paper presents a parameterized algorithmic framework for call graph construction in the presence of message sends and/or first class functions. We use this framework to describe and to implement a number of well-known and new algorithms. We then empirically assess these algorithms by applying them to a suite of medium-sized programs written in Cecil and Java, reporting on the relative cost of the analyses, the relative precision of the constructed call graphs, and the impact of this precision on the effectiveness of a number of interprocedural optimizations.

Previously, techniques such as class hierarchy analysis and profile-guided receiver class prediction have been demonstrated to greatly improve the performance of applications written in pure object-oriented languages, but the degree to which these results are transferable to applications written in hybrid languages has been unclear. In part to answer this question, we have developed the Vortex compiler infrastructure, a language-independent optimizing compiler for object-oriented languages, with front-ends for Cecil, C++, Java, and Modula-3. In this paper, we describe the Vortex compiler's intermediate language, internal structure, and optimization suite, and then we report the results of experiments assessing the effectiveness of different combinations of optimizations on sizable applications across these four languages. We characterize the benchmark programs in terms of a collection of static and dynamic metrics, intended to quantify aspects of the "object-orientedness" of a program.

Compilersand other programming environment tools derive information from the source code of programs; derived information includes compiled code, interprocedurrd summary information, and call graph views.If the source program changes, the derived information needs to be updated.We present a simple framework for maintaining interrnodule dependencies, embodying different tradeoffs in terms of space usage, speed of processing, and selectivity of invalidation, that eases the implementation of incremental update of derived information.Our framework augments a directed acyclic graph representation of dependencies with factoring nodes (to save space) and$ltering nodes (to increase selectivity), and it includes an algorithm for efficient invalidation processing.We show how several schemes for selective recompilation, such as smart recompilation, filter sets for interprocedural summary information, and dependencies for whole-program optimization of object-oriented languages, map naturally onto our framework.For this latter application, by exploiting the facilities of our framework, we are able to reduce the number of lines of source code recompiled by a factor of seven over a header file-based scheme, and by a factor of two over the previous state-of-the-art selective dependency mechanism without consuming additional space.

The use of dynamically-dispatched procedure calls is a key mechanism for writing extensible and flexible code in object-oriented languages. Unfortunately, dynamic dispatching imposes a runtime performance penalty. Some recent implementations of pure object-oriented languages have utilized profile-guided receiver class prediction to reduce this performance penalty, and some researchers have argued for applying receiver class prediction in hybrid languages like C++. We performed a detailed examination of the dynamic profiles of eight large object-oriented applications written in C++ and Cecil, determining that the receiver class distributions are strongly peaked and stable across both inputs and program versions through time. We describe techniques for gathering and manipulating profile information at varying degrees of precision, particularly in the presence of optimizations such as inlining. Our implementation of profile-guided receiver class prediction improves the performance of large Cecil applications by more than a factor of two over solely static optimizations.

Dynamic dispatching is a major source of run-time overhead in object-oriented languages, due both to the direct cost of method lookup and to the indirect effect of preventing other optimizations. To reduce this overhead, optimizing compilers for object-oriented languages analyze the classes of objects stored in program variables, with the goal of bounding the possible classes of message receivers enough so that the compiler can uniquely determine the target of a message send at compile time and replace the message send with a direct procedure call. Specialization is one important technique for improving the precision of this static class information: by compiling multiple versions of a method, each applicable to a subset of the possible argument classes of the method, more precise static information about the classes of the method's arguments is obtained. Previous specialization strategies have not been selective about where this technique is applied, and therefore tended to significantly increase compile time and code space usage, particularly for large applications. In this paper, we present a more general framework for specialization in object-oriented languages and describe a goal directed specialization algorithm that makes selective decisions to apply specialization to those cases where it provides the highest benefit. Our results show that our algorithm improves the performance of a group of sizeable programs by 65% to 275% while increasing compiled code space requirements by only 4% to 10%. Moreover, when compared to the previous state-of-the-art specialization scheme, our algorithm improves performance by 11% to 67% while simultaneously reducing code space requirements by 65% to 73%.