Dynamic adaptive pre-tenuring

Harris, Timothy L.

doi:10.1145/362422.362476

Cited by 41 publications

(17 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Harris [14] uses Agesen and Garthwaite's sampling mechanism to make dynamic pretenuring decisions for Java programs in the context of a two generation collector. When the system detects a long-lived allocation site, it begins allocating into a mature generation.…”

Section: Related Workmentioning

confidence: 99%

“…Lifetime sampling computes transient and total object survival statistics This mechanism samples larger objects more frequently and thus yields more accurate statistics for them. As Harris [14] points out, large objects are important -especially if they are prolific.…”

Section: Object Samplingmentioning

confidence: 99%

“…For the other cases, the overheads and pretenuring errors slow programs down on average by 1% to 4%, and up to 16%. Prior dynamic pretenuring work [14,15] achieves smaller improvements and similar worst case degradations, e.g., 205 raytrace slows down by 15% [14]. To prevent degradations when pretenuring is not applicable, the system could trigger sampling only when nursery survival statistics indicate some potential.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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%

Section: Object Samplingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

“…One approach of them is pretenur-ing [6], [5], [8] . In the approach, the memory manager tries to find out which objects have long lifetimes and allocate such objects directly in the mature space.…”

Section: Related Workmentioning

confidence: 99%

“…This profiling is then used to classify allocation sites. Harris [8] uses dynamic sampling to predict the lifetimes of objects. Sampling reduces the cost of obtaining statistics so that it avoids the need for a separate profiling phase and allows pretenuring decisions changed during runtime.…”

Section: Related Workmentioning

confidence: 99%