The Experts below are selected from a list of 303 Experts worldwide ranked by ideXlab platform

Tu Shi-hang - One of the best experts on this subject based on the ideXlab platform.

  • Research and analysis of Garbage Collection mechanism for real-time embedded Java
    8th International Conference on Computer Supported Cooperative Work in Design, 2004
    Co-Authors: Chen Zhang-long, Tu Shi-hang
    Abstract:

    Real-time embedded software is mostly written in low level language such as C and assembler. But such systems are hard to test, maintain, and port to other platform. The Java language is entering to the real-time embedded systems because of its robustness, simplicity and portability etc. However, It is very difficult to estimate the worst-case execution time of Java code currently. This is due to dynamic class loading and method calls, especially the Automatic Garbage Collection. The paper has research-ed present solutions for Garbage Collection, and studied the memory management of the real-time specification of Java (RTSJ). Then the rational solute-on for RTSJ is to adapt incremental Garbage Collection and the region memory management.

Andreas Moshovos - One of the best experts on this subject based on the ideXlab platform.

  • PACT - Pointy: a hybrid pointer prefetcher for managed runtime systems
    Proceedings of the 21st international conference on Parallel architectures and compilation techniques - PACT '12, 2012
    Co-Authors: Ioana Burcea, Livio Soares, Andreas Moshovos
    Abstract:

    This work proposes Pointy, a software assisted hardware pointer prefetcher for Java applications. Pointy exploits the strengths of both software and hardware. Its runtime software component communicates points-to relationships between objects to the underlying hardware. This points-to information is maintained and tracked in any managed runtime that implements Automatic Garbage Collection. Pointy stores the object connectivity information in a separate hardware structure and uses it to generate timely pointer prefetches. To achieve a low-cost hardware implementation, Pointy spills the object metadata to the conventional memory hierarchy and retrieves it only when needed. Taking advantage of its hybrid design, Pointy can selectively communicate points-to metadata to the hardware based on class profiling that is readily available at the runtime level, while impractical to extract at the hardware level. Experimental results show that Pointy improves performance for pointer intensive benchmarks even in the presence of conventional prefetchers. When used in conjunction with traditional prefetchers, such as striding and next line, Pointy improves application performance by 53% for SpecJBB 2005 and by 72% on average1, which represents a speedup of 19% and 18%, respectively, compared to traditional prefetchers.

  • Pointy: A hybrid pointer prefetcher for managed runtime systems
    2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT), 2012
    Co-Authors: Ioana Burcea, Livio Soares, Andreas Moshovos
    Abstract:

    This work proposes Pointy, a software assisted hardware pointer prefetcher for Java applications. Pointy exploits the strengths of both software and hardware. Its runtime software component communicates points-to relationships between objects to the underlying hardware. This points-to information is maintained and tracked in any managed runtime that implements Automatic Garbage Collection. Pointy stores the object connectivity information in a separate hardware structure and uses it to generate timely pointer prefetches. To achieve a low-cost hardware implementation, Pointy spills the object metadata to the conventional memory hierarchy and retrieves it only when needed. Taking advantage of its hybrid design, Pointy can selectively communicate points-to metadata to the hardware based on class profiling that is readily available at the runtime level, while impractical to extract at the hardware level. Experimental results show that Pointy improves performance for pointer intensive benchmarks even in the presence of conventional prefetchers. When used in conjunction with traditional prefetchers, such as striding and next line, Pointy improves application performance by 53% for SpecJBB 2005 and by 72% on average1, which represents a speedup of 19% and 18%, respectively, compared to traditional prefetchers.

Dennis G. Kafura - One of the best experts on this subject based on the ideXlab platform.

  • Concurrent and distributed Garbage Collection of active objects
    IEEE Transactions on Parallel and Distributed Systems, 1995
    Co-Authors: Dennis G. Kafura, Manibrata Mukherji, Doug Washabaugh
    Abstract:

    This paper shows how to perform concurrent and distributed Automatic Garbage Collection of objects possessing their own thread of control. The relevance of Garbage Collection and active objects to distributed applications is briefly discussed and the specific model of active objects used in the paper is explained. The collector is comprised of independent local collectors, one per node, and a distributed global collector. The mutator (application), the local collectors and the global collector run concurrently. An important part of this paper is the detailed presentation of the algorithms necessary to achieve correct concurrent operation among the collectors and between the collectors and the mutator. The collector builds on previous algorithms for taking snapshots in distributed systems and for detecting termination. >

  • Distributed Garbage Collection of active objects
    [1991] Proceedings. 11th International Conference on Distributed Computing Systems, 1991
    Co-Authors: Doug Washabaugh, Dennis G. Kafura
    Abstract:

    Distributed Automatic Garbage Collection of objects possessing their own thread of control is discussed. The relevance of Garbage Collection and concurrent objects to distributed applications is briefly discussed, and the specific model of concurrent objects used is explained. The collector comprises a Collection of independent local collectors, one per node, loosely coupled to a distributed global collector. The mutator (application), the local collectors, and the global collector run concurrently. The synchronization necessary to achieve correct and efficient concurrent operation between the collectors is presented. One interesting aspect of the distributed collector is the termination algorithm.

  • ICDCS - Distributed Garbage Collection of active objects
    [1991] Proceedings. 11th International Conference on Distributed Computing Systems, 1991
    Co-Authors: Doug Washabaugh, Dennis G. Kafura
    Abstract:

    Distributed Automatic Garbage Collection of objects possessing their own thread of control is discussed. The relevance of Garbage Collection and concurrent objects to distributed applications is briefly discussed, and the specific model of concurrent objects used is explained. The collector comprises a Collection of independent local collectors, one per node, loosely coupled to a distributed global collector. The mutator (application), the local collectors, and the global collector run concurrently. The synchronization necessary to achieve correct and efficient concurrent operation between the collectors is presented. One interesting aspect of the distributed collector is the termination algorithm. >

  • RTSS - Incremental Garbage Collection of concurrent objects for real-time applications
    [1990] Proceedings 11th Real-Time Systems Symposium, 1990
    Co-Authors: Doug Washabaugh, Dennis G. Kafura
    Abstract:

    The authors show how to perform real-time Automatic Garbage Collection of objects possessing their own thread of control. Beyond its interest as a novel real-time problem, the relevance of Automatic management and concurrent objects to real-time applications is briefly discussed. The specific model of concurrent objects used in the paper is explained. A definition of real-time Garbage Collection is given and an algorithm satisfying this definition is described. An analysis of the relationship between latency, memory overhead and system size shows that this approach is immediately feasible for low-performance real-time systems or multiprocessor real-time systems with a dedicated processor for Garbage Collection. Future improvements in the collector's performance are outlined. >

  • Real-Time Garbage Collection of Actors
    1990
    Co-Authors: Doug Washabaugh, Dennis G. Kafura
    Abstract:

    This paper shows how to perform real-time Automatic Garbage Collection of objects possessing their own thread of control. Beyond its interest as a novel real-time problem, the relevance of Automatic management and concurrent objects to real-time applications is briefly discussed. The specific model of concurrent objects used in the paper is explained. A definition of real-time Garbage Collection is given and an algorithm satisfying this definition is described. An analysis of the relationship between latency, memory overhead and system size shows that this approach is immediately feasible for low-performance real-time systems or multi-processor real-time systems with dedicated processor functions. Future improvements in the collector''s performance are outlined.

Chen Zhang-long - One of the best experts on this subject based on the ideXlab platform.

  • Research and analysis of Garbage Collection mechanism for real-time embedded Java
    8th International Conference on Computer Supported Cooperative Work in Design, 2004
    Co-Authors: Chen Zhang-long, Tu Shi-hang
    Abstract:

    Real-time embedded software is mostly written in low level language such as C and assembler. But such systems are hard to test, maintain, and port to other platform. The Java language is entering to the real-time embedded systems because of its robustness, simplicity and portability etc. However, It is very difficult to estimate the worst-case execution time of Java code currently. This is due to dynamic class loading and method calls, especially the Automatic Garbage Collection. The paper has research-ed present solutions for Garbage Collection, and studied the memory management of the real-time specification of Java (RTSJ). Then the rational solute-on for RTSJ is to adapt incremental Garbage Collection and the region memory management.

Ioana Burcea - One of the best experts on this subject based on the ideXlab platform.

  • PACT - Pointy: a hybrid pointer prefetcher for managed runtime systems
    Proceedings of the 21st international conference on Parallel architectures and compilation techniques - PACT '12, 2012
    Co-Authors: Ioana Burcea, Livio Soares, Andreas Moshovos
    Abstract:

    This work proposes Pointy, a software assisted hardware pointer prefetcher for Java applications. Pointy exploits the strengths of both software and hardware. Its runtime software component communicates points-to relationships between objects to the underlying hardware. This points-to information is maintained and tracked in any managed runtime that implements Automatic Garbage Collection. Pointy stores the object connectivity information in a separate hardware structure and uses it to generate timely pointer prefetches. To achieve a low-cost hardware implementation, Pointy spills the object metadata to the conventional memory hierarchy and retrieves it only when needed. Taking advantage of its hybrid design, Pointy can selectively communicate points-to metadata to the hardware based on class profiling that is readily available at the runtime level, while impractical to extract at the hardware level. Experimental results show that Pointy improves performance for pointer intensive benchmarks even in the presence of conventional prefetchers. When used in conjunction with traditional prefetchers, such as striding and next line, Pointy improves application performance by 53% for SpecJBB 2005 and by 72% on average1, which represents a speedup of 19% and 18%, respectively, compared to traditional prefetchers.

  • Pointy: A hybrid pointer prefetcher for managed runtime systems
    2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT), 2012
    Co-Authors: Ioana Burcea, Livio Soares, Andreas Moshovos
    Abstract:

    This work proposes Pointy, a software assisted hardware pointer prefetcher for Java applications. Pointy exploits the strengths of both software and hardware. Its runtime software component communicates points-to relationships between objects to the underlying hardware. This points-to information is maintained and tracked in any managed runtime that implements Automatic Garbage Collection. Pointy stores the object connectivity information in a separate hardware structure and uses it to generate timely pointer prefetches. To achieve a low-cost hardware implementation, Pointy spills the object metadata to the conventional memory hierarchy and retrieves it only when needed. Taking advantage of its hybrid design, Pointy can selectively communicate points-to metadata to the hardware based on class profiling that is readily available at the runtime level, while impractical to extract at the hardware level. Experimental results show that Pointy improves performance for pointer intensive benchmarks even in the presence of conventional prefetchers. When used in conjunction with traditional prefetchers, such as striding and next line, Pointy improves application performance by 53% for SpecJBB 2005 and by 72% on average1, which represents a speedup of 19% and 18%, respectively, compared to traditional prefetchers.