Learned Garbage Collection

Simple item page
dc.contributor.authorCen, Lujing
dc.contributor.authorMarcus, Ryan
dc.contributor.authorGottschlich, Justin E
dc.contributor.authorGottschlich, Justin E
dc.contributor.authorAlizadeh, Mohammad
dc.contributor.authorKraska, Tim
dc.date2023-05-18T00:14:26.000
dc.date.accessioned2023-05-22T13:06:48Z
dc.date.available2023-05-22T13:06:48Z
dc.date.issued2020-01-01
dc.date.submitted2020-12-18T13:22:47-08:00
dc.description.abstractSeveral programming languages use garbage collectors (GCs) to automatically manage memory for the programmer. Such collectors must decide when to look for unreachable objects to free, which can have a large performance impact on some applications. In this preliminary work, we propose a design for a learned garbage collector that autonomously learns over time when to perform collections. By using reinforcement learning, our design can incorporate user-defined reward functions, allowing an autonomous garbage collector to learn to optimize the exact metric the user desires (e.g., request latency or queries per second). We conduct an initial experimental study on a prototype, demonstrating that an approach based on tabular Q learning may be promising.
dc.identifier.urihttps://repository.upenn.edu/handle/20.500.14332/8492
dc.legacy.articleid1005
dc.legacy.fulltexturlhttps://repository.upenn.edu/cgi/viewcontent.cgi?article=1005&context=cps_machine_programming&unstamped=1
dc.source.issue6
dc.source.journalMachine Programming
dc.source.statuspublished
dc.titleLearned Garbage Collection
dc.typePresentation
digcom.contributor.authorCen, Lujing
digcom.contributor.authorMarcus, Ryan
digcom.contributor.authorMao, Hongzi
digcom.contributor.authorisAuthorOfPublication|email:gojustin@cis.upenn.edu|institution:Intel|Gottschlich, Justin E
digcom.contributor.authorAlizadeh, Mohammad
digcom.contributor.authorKraska, Tim
digcom.identifiercps_machine_programming/6
digcom.identifier.contextkey20688903
digcom.identifier.submissionpathcps_machine_programming/6
digcom.typeconference
dspace.entity.typePublication
relation.isAuthorOfPublication5cbcf403-a558-4c1c-aa8a-d700e3d50679
relation.isAuthorOfPublication5cbcf403-a558-4c1c-aa8a-d700e3d50679
relation.isAuthorOfPublication.latestForDiscovery5cbcf403-a558-4c1c-aa8a-d700e3d50679
upenn.schoolDepartmentCenterMachine Programming
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2004.13301.pdf
Size:
363.83 KB
Format:
Adobe Portable Document Format
Download
Collection
Presentations