Kathryn S. McKinley

The DaCapo benchmarks: java benchmarking development and analysis

Since benchmarks drive computer science research and industry product development, which ones we use and how we evaluate them are key questions for the community. Despite complex runtime tradeoffs due to dynamic compilation and garbage collection required for Java programs, many evaluations still use methodologies developed for C, C++, and Fortran. SPEC, the dominant purveyor of benchmarks, compounded this problem by institutionalizing these methodologies for their Java benchmark suite. This paper recommends benchmarking selection and evaluation methodologies, and introduces the DaCapo benchmarks, a set of open source, client-side Java benchmarks. We demonstrate that the complex interactions of (1) architecture, (2) compiler, (3) virtual machine, (4) memory management, and (5) application require more extensive evaluation than C, C++, and Fortran which stress (4) much less, and do not require (3). We use and introduce new value, time-series, and statistical metrics for static and dynamic properties such as code complexity, code size, heap composition, and pointer mutations. No benchmark suite is definitive, but these metrics show that DaCapo improves over SPEC Java in a variety of ways, including more complex code, richer object behaviors, and more demanding memory system requirements. This paper takes a step towards improving methodologies for choosing and evaluating benchmarks to foster innovation in system design and implementation for Java and other managed languages.

Improving data locality with loop transformations

In the past decade, processor speed has become significantly faster than memory speed. Small, fast cache memories are designed to overcome this discrepancy, but they are only effective when programs exhibit data locality. In the this article, we present compiler optimizations to improve data locality based on a simple yet accurate cost model. The model computes both temporal and spatial reuse of cache lines to find desirable loop organizations. The cost model drives the application of compound transformations consisting of loop permutation, loop fusion, loop distribution, and loop reversal. To validate our optimization strategy, we implemented our algorithms and ran experiments on a large collection of scientific programs and kernels. Experiments illustrate that for kernels our model and algorithm can select and achieve the best loop structure for a nest. For over 30 complete applications, we executed the original and transformed versions and simulated cache hit rates. We collected statistics about the inherent characteristics of these programs and our ability to improve their data locality. To our knowledge, these studies are the first of such breadth and depth. We found performance improvements were difficult to achieve bacause benchmark programs typically have high hit rates even for small data caches; however, our optimizations significanty improved several programs.

Hoard: a scalable memory allocator for multithreaded applications

Parallel, multithreaded C and C++ programs such as web servers, database managers, news servers, and scientific applications are becoming increasingly prevalent. For these applications, the memory allocator is often a bottleneck that severely limits program performance and scalability on multiprocessor systems. Previous allocators suffer from problems that include poor performance and scalability, and heap organizations that introduce false sharing. Worse, many allocators exhibit a dramatic increase in memory consumption when confronted with a producer-consumer pattern of object allocation and freeing. This increase in memory consumption can range from a factor of P (the number of processors) to unbounded memory consumption.This paper introduces Hoard, a fast, highly scalable allocator that largely avoids false sharing and is memory efficient. Hoard is the first allocator to simultaneously solve the above problems. Hoard combines one global heap and per-processor heaps with a novel discipline that provably bounds memory consumption and has very low synchronization costs in the common case. Our results on eleven programs demonstrate that Hoard yields low average fragmentation and improves overall program performance over the standard Solaris allocator by up to a factor of 60 on 14 processors, and up to a factor of 18 over the next best allocator we tested.

Tile size selection using cache organization and data layout

When dense matrix computations are too large to fit in cache, previous research proposes tiling to reduce or eliminate capacity misses. This paper presents a new algorithm for choosing problem-size dependent tile sizes based on the cache size and cache line size for a direct-mapped cache. The algorithm eliminates both capacity and self-interference misses and reduces cross-interference misses. We measured simulated miss rates and execution times for our algorithm and two others on a variety of problem sizes and cache organizations. At higher set associativity, our algorithm does not always achieve the best performance. However on direct-mapped caches, our algorithm improves simulated miss rates and measured execution times when compared with previous work.

Scaling to the end of silicon with EDGE architectures

Microprocessor designs are on the verge of a post-RISC era in which companies must introduce new ISAs to address the challenges that modern CMOS technologies pose while also exploiting the massive levels of integration now possible. To meet these challenges, we have developed a new class of ISAs, called explicit data graph execution (EDGE), that will match the characteristics of semiconductor technology over the next decade. The TRIPS architecture is the first instantiation of an EDGE instruction set, a new, post-RISC class of instruction set architectures intended to match semiconductor technology evolution over the next decade, scaling to new levels of power efficiency and high performance.

Compiler optimizations for improving data locality

In the past decade, processor speed has become significantly faster than memory speed. Small, fast cache memories are designed to overcome this discrepancy, but they are only effective when programs exhibit data locality. In this paper, we present compiler optimizations to improve data locality based on a simple yet accurate cost model. The model computes both temporal and spatial reuse of cache lines to find desirable loop organizations. The cost model drives the application of compound transformations consisting of loop permutation, loop fusion, loop distribution, and loop reversal. We demonstrate that these program transformations are useful for optimizing many programs. To validate our optimization strategy, we implemented our algorithms and ran experiments on a large collection of scientific programs and kernels. Experiments with kernels illustrate that our model and algorithm can select and achieve the best performance. For over thirty complete applications, we executed the original and transformed versions and simulated cache hit rates. We collected statistics about the inherent characteristics of these programs and our ability to improve their data locality. To our knowledge, these studies are the first of such breadth and depth. We found performance improvements were difficult to achieve because benchmark programs typically have high hit rates even for small data caches; however, our optimizations significantly improved several programs.

Maximizing Loop Parallelism and Improving Data Locality via Loop Fusion and Distribution

Loop fusion is a program transformation that merges multiple loops into one. It is effective for reducing the synchronization overhead of parallel loops and for improving data locality. This paper presents three results for fusion: (1) a new algorithm for fusing a collection of parallel and sequential loops, minimizing parallel loop synchronization while maximizing parallelism; (2) a proof that performing fusion to maximize data locality is NP-hard; and (3) two polynomial-time algorithms for improving data locality. These techniques also apply to loop distribution, which is shown to be essentially equivalent to loop fusion. Our approach is general enough to support other fusion heuristics. Preliminary experimental results validate our approach for improving performance by exploiting data locality and increasing the granularity of parallelism.

The Jikes research virtual machine project: building an open-source research community

This paper describes the evolution of the JikesTM Research Virtual Machine project from an IBM internal research project, called Jalapeno, into an open-source project. After summarizing the original goals of the project, we discuss the motivation for releasing it as an open-source project and the activities performed to ensure the success of the project. Throughout, we highlight the unique challenges of developing and maintaining an open-source project designed specifically to support a research community.

Oil and water? High performance garbage collection in Java with MMTk

Increasingly popular languages such as Java and C# require efficient garbage collection. This paper presents the design, implementation, and evaluation of MMTk, a Memory Management Toolkit for and in Java. MMTk is an efficient, composable, extensible, and portable framework for building garbage collectors. MMTk uses design patterns and compiler cooperation to combine modularity and efficiency. The resulting system is more robust, easier to maintain, and has fewer defects than monolithic collectors. Experimental comparisons with monolithic Java and C implementations reveal MMTk has significant performance advantages as well. Performance critical system software typically uses monolithic C at the expense of flexibility. Our results refute common wisdom that only this approach attains efficiency, and suggest that performance critical software can embrace modular design and high-level languages.

Reconsidering custom memory allocation

Programmers hoping to achieve performance improvements often use custom memory allocators. This in-depth study examines eight applications that use custom allocators. Surprisingly, for six of these applications, a state-of-the-art general-purpose allocator (the Lea allocator) performs as well as or better than the custom allocators. The two exceptions use regions, which deliver higher performance (improvements of up to 44%). Regions also reduce programmer burden and eliminate a source of memory leaks. However, we show that the inability of programmers to free individual objects within regions can lead to a substantial increase in memory consumption. Worse, this limitation precludes the use of regions for common programming idioms, reducing their usefulness.We present a generalization of general-purpose and region-based allocators that we call reaps. Reaps are a combination of regions and heaps, providing a full range of region semantics with the addition of individual object deletion. We show that our implementation of reaps provides high performance, outperforming other allocators with region-like semantics. We then use a case study to demonstrate the space advantages and software engineering benefits of reaps in practice. Our results indicate that programmers needing fast regions should use reaps, and that most programmers considering custom allocators should instead use the Lea allocator.