Vectorized garbage collection

Appel, Andrew W.; Bendiksen, Aage

doi:10.1007/bf00127826

Cited by 15 publications

(28 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hash consing merges equivalent objects [4], and prior work uses hash consing to compact dynamic call trees [29]. The CCU-and CCT-based approaches are essentially performing hash consing of nodes: the CCT performs hash consing eagerly, and the CCU performs hash consing lazily.…”

Section: Quantitatively Evaluating Context-sensitive Racy Sitesmentioning

confidence: 99%

Efficient context sensitivity for dynamic analyses via calling context uptrees and customized memory management

Huang

Bond

2013

Proceedings of the 2013 ACM SIGPLAN International Conference on Object Oriented Programming Systems Languages &Amp; Application

View full text Add to dashboard Cite

State-of-the-art dynamic bug detectors such as data race and memory leak detectors report program locations that are likely causes of bugs. However, programmers need more than static program locations to understand the behavior of increasingly complex and concurrent software. Dynamic calling context provides additional information, but it is expensive to record calling context frequently, e.g., at every read and write. Context-sensitive dynamic analyses can build and maintain a calling context tree (CCT) to track calling context-but in order to reuse existing nodes, CCT-based approaches require an expensive lookup. This thesis introduces a new approach for context sensitivity that avoids this expensive lookup. The approach uses a new data structure called the calling context I would like to start by giving my deepest gratitude to my advisor Mike Bond, who has spent a tremendous amount of time and energy trying to train me into a good researcher. Since Jan 2011, I have been happy working with him and he has taught me a lot not only about technical skills but also the way to work efficiently. His unwavering patience and confidence in me have been a constant source of encouragement that helped me overcome numerous challenges and difficulties, and keep making my project as perfect as possible. During these past 3 years, I have acquired more technical skills and confidence with his guidance. Especially when I started working with him, he gave a lot of thoughtful suggestions about course training, research and personal growth.

show abstract

Section: Quantitatively Evaluating Context-sensitive Racy Sitesmentioning

confidence: 99%

Efficient context sensitivity for dynamic analyses via calling context uptrees and customized memory management

Huang

Bond

2013

Proceedings of the 2013 ACM SIGPLAN International Conference on Object Oriented Programming Systems Languages &Amp; Application

View full text Add to dashboard Cite

show abstract

“…The following table shows the cumulative sum of heap sizes across all collections within a run, as well as the equivalent numbers for the reference graph. 4 This allows us to estimate that the reference graph approach reduces the size of our graph by about 75% on average: We experimented with two different approaches to building and maintaining the reference graph. Both of them allocate a node in the reference graph whenever a new object is allocated.…”

Section: Reference Graphmentioning

confidence: 99%

“…Barabash and Petrank cover the problem of garbage collection on highly parallel platforms from a more general perspective and perform a heap analysis similar to ours [5]. An early paper by Appel and Bendiksen [4] covers garbage collection on vector processors and our approach has been influenced by some of their ideas.…”

Section: Related Workmentioning

confidence: 99%

GPUs as an opportunity for offloading garbage collection

Maas

Reames

Morlan

et al. 2012

Proceedings of the 2012 International Symposium on Memory Management

View full text Add to dashboard Cite

GPUs have become part of most commodity systems. Nonetheless, they are often underutilized when not executing graphicsintensive or special-purpose numerical computations, which are rare in consumer workloads. Emerging architectures, such as integrated CPU/GPU combinations, may create an opportunity to utilize these otherwise unused cycles for offloading traditional systems tasks. Garbage collection appears to be a particularly promising candidate for offloading, due to the popularity of managed languages on consumer devices.We investigate the challenges for offloading garbage collection to a GPU, by examining the performance trade-offs for the mark phase of a mark & sweep garbage collector. We present a theoretical analysis and an algorithm that demonstrates the feasibility of this approach. We also discuss a number of algorithmic design trade-offs required to leverage the strengths and capabilities of the GPU hardware. Our algorithm has been integrated into the Jikes RVM and we present promising performance results.

show abstract

“…In practice, hash-consing can be expensive because of large memory demands and interaction with garbage collection. In fact, several researchers have argued that hash-consing is too expensive for practical purposes [Pugh, 1988, Appel and Gonçalves, 1993, Murphy et al, 2002. As an alternative to hash consing, Pugh proposed lazy structure sharing [Pugh, 1988].…”

Section: Equalitymentioning

confidence: 99%

Selective memoization

2003

View full text Add to dashboard Cite

This paper presents language techniques for applying memoization selectively. The techniques provide programmer control over equality, space usage, and identification of precise dependences so that memoization can be applied according to the needs of an application. Two key properties of the approach are that it accepts and efficient implementation and yields programs whose performance can be analyzed using standard analysis techniques.We describe our approach in the context of a functional language called MFL and an implementation as a Standard ML library. The MFL language employs a modal type system to enable the programmer to express programs that reveal their true data dependences when executed. We prove that the MFL language is sound by showing that that MFL programs yield the same result as they would with respect to a standard, non-memoizing semantics. The SML implementation cannot support the modal type system of MFL statically but instead employs run-time checks to ensure correct usage of primitives.

show abstract

Vectorized garbage collection

Cited by 15 publications

References 5 publications

Efficient context sensitivity for dynamic analyses via calling context uptrees and customized memory management

Efficient context sensitivity for dynamic analyses via calling context uptrees and customized memory management

GPUs as an opportunity for offloading garbage collection

Selective memoization

Contact Info

Product

Resources

About