Exploiting frequent field values in java objects for reducing heap memory requirements

Chen, Guangyu; Kandemir, Mahmut; Irwin, M.J.

doi:10.1145/1064979.1064990

Cited by 10 publications

(13 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This technique requires two objects to be immutable in order to be merged into an identical one. Eliminating unused fields in objects to save space has been studied by Ananian and Rinard and Chen et al [4,6]. Our mechanism improves upon theirs by better exploiting the space-saving opportunities associated with unused fields.…”

Section: Related Workmentioning

confidence: 98%

“…Chen, et al [6] present a detailed characterization of frequent field values in the SpecJVM98 benchmark and propose two heap compression schemes. They used a heuristic formula to classify the fields of a class into three levels: At Level-0 the field does not have a dominant frequent field value; at Level-1 the field has a non-zero frequent field value; and at Level-2 the field has a frequent field value that is zero or null.…”

Section: Previous Use Of Field Usage Patternsmentioning

confidence: 99%

“…Like Chen et al [6], our field compression mechanism also classifies fields into three levels: level-0 fields have no frequent field values, level-1 fields have non-zero frequent field values and level-2 fields have zero (or null) frequent field values. However, instead of using an heuristic formula to predict the level of a field, we use profiling information.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…The run-time overhead to access compressed objects is estimated by counting the number of ex- we assume that each instruction is executed in one machine cycle and that each memory access requires an extra cycle [6]. They argue that this assumption is reasonable for embedded microprocessors with low clock frequencies and short pipeline architectures.…”

Section: Performance Overheadmentioning

confidence: 99%

“…Several researchers have attempted to reduce the data footprint of Java programs by eliminating unused fields [6,19,4,20]. However, none of them have fully exploited the opportunities for space optimization that can be obtained by studying field usage patterns to identify unused fields.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Utilizing field usage patterns for Java heap space optimization

Guo

Amaral

Szafron

et al. 2006

Proceedings of the 2006 Conference of the Center for Advanced Studies on Collaborative Research - CASCON '06

View full text Add to dashboard Cite

This research studies the characteristics of field usage patterns in the SpecJVM98 benchmarks. It finds that multiple object instances of the same class often exhibit different field-usage patterns. Motivated by this observation, we designed a heap compression mechanism that classifies object instances at runtime based on their field-usage patterns and eliminates unused fields to save space. To achieve the maximum space savings while minimizing the space and time overhead, our design combines three interrelated techniques in a novel manner: runtime object instance classification, field virtualization, and bidirectional object layout. An experimental evaluation reveals that this mechanism can reduce the maximum heap occupancy of SpecJVM98 benchmarks by up to 18% and 14% on average while keeping the application execution overhead low.

show abstract

Section: Related Workmentioning

confidence: 98%

Section: Previous Use Of Field Usage Patternsmentioning

confidence: 99%

Section: Implementation Detailsmentioning

confidence: 99%

Section: Performance Overheadmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Utilizing field usage patterns for Java heap space optimization

Guo

Amaral

Szafron

et al. 2006

Proceedings of the 2006 Conference of the Center for Advanced Studies on Collaborative Research - CASCON '06

View full text Add to dashboard Cite

show abstract

Object-Relative Addressing: Compressed Pointers in 64-Bit Java Virtual Machines

Venstermans

Eeckhout

Bosschere

2007

ECOOP 2007 – Object-Oriented Programming

View full text Add to dashboard Cite

EMPLOYMENT CHARACTERISTICS OF FAMILIES -2017In 2017, 5.8 percent of families included an unemployed person, down from 6.5 percent in 2016, the U.S. Bureau of Labor Statistics reported today. Of the nation's 82.0 million families, 80.5 percent had at least one employed member in 2017.These data on employment, unemployment, and family relationships are collected as part of the Current Population Survey (CPS), a monthly survey of about 60,000 households. Data in this release are annual averages. Families are classified either as married-couple families or as families maintained by women or men without spouses present. Unless otherwise noted, families include those with and without children under age 18. For further information, see the Technical Note in this news release. Families and UnemploymentThe number of families with at least one member unemployed decreased by 557,000 to 4.7 million in 2017. The proportion of families with an unemployed person declined by 0.7 percentage point to 5.8 percent. In 2017, this proportion was down for White (5.2 percent), Black (9.5 percent), and Hispanic (7.7 percent) families and was little changed for Asian families (5.4 percent In 2017, 4.6 percent of married-couple families had an unemployed member, less than the corresponding percentages of families maintained by women or families maintained by men (9.0 percent and 8.6 percent, respectively). Among families with an unemployed family member, those maintained by women were less likely to also have an employed family member (49.8 percent) than families maintained by men and married-couple families (59.4 percent and 80.7 percent, respectively).(See tables 2 and 3.)-2- Families and EmploymentIn 2017, 80.5 percent of families had at least one employed family member, little different from the prior year. Over the year, the likelihood of having an employed family member was about unchanged among White (80.1 percent), Asian (88.6 percent), and Hispanic families (86.9 percent). Families with ChildrenIn 2017, 33.6 million families included children under age 18, about 41.0 percent of all families.(Children are sons, daughters, step-children, or adopted children living in the household who are under 18 years old. Not included are nieces, nephews, grandchildren, other related and unrelated children, and children not living in the household.) At least one parent was employed in 90.2 percent of families with children, an increase of 0.5 percentage point from the previous year. Among married-couple families with children, 96.9 percent had at least one employed parent, and 61.9 percent had both parents employed. Among families with children, 84.7 percent of fathers were employed in those maintained by fathers, a greater share than the 73.2 percent of mothers who were employed in families maintained by mothers. (See table 4.) ParentsThe labor force participation rate-the percent of the population working or looking for work-for all women with children under age 18 was 71.1 percent in 2017, up 0.6 percentage point from the prior year. Married...

show abstract