Generational garbage collection and the radioactive decay model

Clinger, William D.; Hansen, Lars T.

doi:10.1145/258916.258925

Cited by 12 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other than the weak generational hypothesis [16,25], previous work using analytical modeling and experimental classification across programs has not yielded any additional general object lifetime hypotheses [9,22]. However, many memory management techniques improve performance for a given program based on its individual characteristics.…”

Section: Related Workmentioning

confidence: 99%

Dynamic object sampling for pretenuring

Jump

Blackburn

McKinley

2004

Proceedings of the 4th International Symposium on Memory Management

View full text Add to dashboard Cite

Many state-of-the-art garbage collectors are generational, collecting the young nursery objects more frequently than old objects. These collectors perform well because young objects tend to die at a higher rate than old ones. However, these collectors do not examine object lifetimes with respect to any particular program or allocation site. This paper introduces low-cost object sampling to dynamically determine lifetimes. The sampler marks an object and records its allocation site every n bytes of allocation. The collector then computes per-site nursery survival rates. Sampling degrades total performance by only 3% on average for sample rates of 256 bytes in Jikes RVM, a rate at which overall lifetime accuracy compares well with sampling every object.An adaptive collector can use this information to tune itself. For example, pretenuring decreases nursery collection work by allocating new, but long-lived, objects directly into the mature space. We introduce a dynamic pretenuring mechanism that detects long-lived allocation sites and pretenures them, given sufficient samples. To react to phase changes, it occasionally backsamples. As with previous online pretenuring, consistent performance improvements on SPECjvm98 benchmarks are difficult to attain since only two combine sufficient allocation load with high nursery survival. Our pretenuring system consistently improves one of these, 213 javac, by 2% to 9% of total time by decreasing collection time by over a factor of two. Sampling and pretenuring overheads slow down all the others. This paper thus provides an efficient sampling mechanism that accurately predicts lifetimes, but leaves open optimization policies that can exploit this information.

show abstract

Section: Related Workmentioning

confidence: 99%

Dynamic object sampling for pretenuring

Jump

Blackburn

McKinley

2004

Proceedings of the 4th International Symposium on Memory Management

View full text Add to dashboard Cite

show abstract

“…Baker offered a radioactive decay model of object lifetimes as an example of a distribution that cannot benefit from generational garbage collection. Although Baker expected a generational collector to perform neither better nor worse than a non-generation collector on this model, Clinger and Hansen calculated that younger-first generational collection would perform worse [11]. Clinger and Rojas argued, however, that a linear combination of these models (for young, intermediate and older objects) was adequate to model simple generation collectors [12].…”

Section: Related Workmentioning

confidence: 96%

The economics of garbage collection

Singer

Jones

Brown

et al. 2010

Proceedings of the 2010 International Symposium on Memory Management

View full text Add to dashboard Cite

This paper argues that economic theory can improve our understanding of memory management. We introduce the allocation curve, as an analogue of the demand curve from microeconomics. An allocation curve for a program characterises how the amount of garbage collection activity required during its execution varies in relation to the heap size associated with that program. The standard treatment of microeconomic demand curves (shifts and elasticity) can be applied directly and intuitively to our new allocation curves. As an application of this new theory, we show how allocation elasticity can be used to control the heap growth rate for variable sized heaps in Jikes RVM.

show abstract

“…Clinger and Hansen [8] explore this radioactive half-life analogy further. They present a meta-analysis to suggest that long-lived objects have lifetimes that are independent of age (i.e.…”

Section: Application Of Mathematical Frameworkmentioning

confidence: 99%

We are all economists now

Cameron

Singer

2014

Proceedings of the 9th International Workshop on Implementation, Compilation, Optimization of Object-Oriented Languages, Progra

View full text Add to dashboard Cite

Multiple virtual machine (VM) workloads are increasingly common, given the growth of managed enterprise application systems and consolidated virtual servers. Until now, there has been no principled approach to partitioning memory resource between multiple co-located VMs. In this paper, we develop a general framework for multi-VM heap sizing, based on the principle of utility maximization from microeconomic theory. We apply utility maximization to static heap sizing, and obtain performance improvements slightly better than current best-practice static heap sizing, and comparable with HotSpot ergonomics (current best-practice dynamic heap sizing). The major advantage of our approach is its simplicity and predictable resource utilization.

show abstract

Generational garbage collection and the radioactive decay model

Cited by 12 publications

References 30 publications

Dynamic object sampling for pretenuring

Dynamic object sampling for pretenuring

The economics of garbage collection

We are all economists now

Contact Info

Product

Resources

About