G. E. Blelloch and P. Cheng. On bounding time and space for multiprocessor garbage collection. In Proceedings of the ACM SIGPLAN '99 Conference on Programming Language Design and Implementation (PLDI), pages 104--117. ACM Press, 1999. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Implementation techniques for a multi-service Java Virtual Machine - Harris (1999) (Correct)
....free objects 3 . Unfortunately, proving the correctness of an untrusted garbage collection policy is beyond the power of current automated techniques. For example, the correctness proofs of proposed garbage collection techniques are generally based on arguments written in natural language [BC99] The intention here is therefore to allow some aspects of garbage collection to be controlled by an application while requiring that the basic collection mechanism is implemented within the JVM or by a trusted programmer. The approach taken is based on using the type information present in the ....
Guy E Blelloch and Perry Cheng. On bounding time and space for multiprocessor garbage collection. ACM SIGPLAN Notices, 34(5):104--117, May 1999.
Using Page Residency to Balance Tradeoffs - In Tracing Garbage Self-citation (Blelloch) (Correct)
No context found.
G. E. Blelloch and P. Cheng. On bounding time and space for multiprocessor garbage collection. In Proceedings of the ACM SIGPLAN '99 Conference on Programming Language Design and Implementation (PLDI), pages 104--117. ACM Press, 1999.
Scalable Room Synchronizations - Blelloch, Cheng, Gibbons (2003) Self-citation (Blelloch Cheng) (Correct)
No context found.
G. E. Blelloch and P. Cheng. On bounding time and space for multiprocessor garbage collection. In Proceedings of the ACM SIGPLAN '99 Conference on Programming Languages Design and Implementation, pages 104--117, May 1999.
Scalable Real-time Parallel Garbage Collection for Symmetric.. - Cheng (2001) (1 citation) Self-citation (Cheng) (Correct)
....Copy, Mark Sweep Larose, Feeley [49] 5 15 ms write barrier Mark Sweep Lim al. 51] ms write barrier, VM support Treadmill Huelsbergen [42] 80 ms write barrier Mark Sweep Endo [26] Bacon al. 01 2.7 ms write barrier Ref. Count Blelloch, Cheng [12] 99 ffl 0.8 1.8ms barrier Rep. Copy Table 1.1: A feature comparison of some collectors. Conservative and distributed collectors are not included. Closely related collectors are grouped into the same row. An open circle (ffi) indicates only partial support. ....
....and analysis of the algorithms. As a review of copying collectors and for illustration of the model, Chapter 3 presents three copying collectors in the abstract model and compares them. Chapter 4 introduces a simple parallel, concurrent, real time algorithm for the simple abstract model [12]. Proofs of time, space, and correctness are presented. The chapter ends with a discussion of how to obtain a parallel, non concurrent collector. To bridge the gap between the simple model and an actual implementation environment, Chapter 5 will present a more realistic model adding features such ....
Guy E. Blelloch and Perry Cheng. On bounding time and space for multiprocessor garbage collection. In PLDI [61], pages 104--117.
A Parallel, Real-Time Garbage Collector - Cheng, Blelloch (2001) (1 citation) Self-citation (Blelloch Cheng) (Correct)
....15213 fpscheng,guybg cs.cmu.edu ABSTRACT We describe a parallel, real time garbage collector and present experimental results that demonstrate good scalability and good real time bounds. The collector is designed for sharedmemory multiprocessors and is based on an earlier collector algorithm [2], whichprovided fixed bounds on the time any thread must pause for collection. However, since our earlier algorithm was designed for simple analysis, it had some impractical features. This paper presents the extensions necessary for a practical implementation: reducing excessive interleaving, ....
....collectors run a single collector thread concurrentlywithone or more application threads [23, 8] These collectors, however, do not consider running multiple collector threads in parallel. In an earlier paper, we presented a collector algorithm that is both scalably parallel and real time [2]. By making all aspects of the collector incremental (e.g. incrementally copying arrays) and allowing an arbitrary number of application and collector threads to run in parallel, wewere able to achieve tight theoretical bounds on the pause time for any application thread as well as bound the ....
[Article contains additional citation context not shown here]
Guy E. Blelloch and Perry Cheng. On bounding time and space for multiprocessor garbage collection. In Proc. ACM SIGPLAN ConferenceonProgramming Languages Design and Implementation,ACM SIGPLAN Notices, pages 104--117, Atlanta, May 99. ACM Press.
Room Synchronizations - Blelloch, Cheng, Gibbons (2001) Self-citation (Blelloch Cheng) (Correct)
....that the technique can be extended to other data structures such as stacks. Our goal is to develop data structures for handling asynchronous parallel accesses that (a) are linearizable, and (b) do not sequentialize access. The problem arose in the context of a real time parallel garbage collector [3, 5]. In such contexts, whenever one of the program threads allocates memory, it also participates in the garbage collection. Thus the threads access the collector data structures asynchronously, in parallel, and somewhat unpredictably. Linearizability is needed for correctness. Avoiding sequential ....
....Room synchronization can be viewed as extracting from these algorithms (and other previous algorithms using fetch and add) a synchronization construct useful for many problems. In our earlier work on garbage collection we briefly described another weaker version of the room synchronizations [3]. That version had a couple important disadvantages over the version described in this paper. Firstly, it did not guarantee properties P5 P10, unless you assumed certain properties of the exact time taken by each instruction. In particular one process could be starved for an arbitrary amount of ....
G. E. Blelloch and P. Cheng. On bounding time and space for multiprocessor garbage collection. In Proc. ACM SIGPLAN'99 Conf. on Programming Languages Design and Implementation, pages 104--117, May 1999.
Bounded Parallel Garbage Collection: Implementation.. - Vaughan.. Self-citation (Blelloch Cheng) (Correct)
....languages in real time systems. Whilst some garbage collectors do provide bounded execution guarantees for single processor systems [Bak78] algorithms that can provide such guarantees for multiprocessor systems have only recently been forthcoming. A new algorithm proposed by Blelloch and Cheng [BC99] is one such example. The basics of garbage collection operation can be described simply. A program is executed by a mutator, which reads and writes existing objects, and may dynamically request the allocation of space for the storage of new objects. The execution environment possesses a number ....
Blelloch, G.E. & Cheng, P. "On Bounding Time and Space for Multiprocessor Garbage Collection". ACM SIGPLAN Notices 34, 5. Proc. ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), Atlanta, Georgia (1999) pp 104-117.
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