Vigil-KV: Hardware-Software Co-Design to Integrate Strong Latency Determinism into Log-Structured Merge Key-Value Stores

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 173
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorKwon, Miryeongko
dc.contributor.authorLee, Seungjunko
dc.contributor.authorChoi, Hyunkyuko
dc.contributor.authorHwang, Jooyoungko
dc.contributor.authorJung, Myoungsooko
dc.date.accessioned2022-12-05T01:01:56Z-
dc.date.available2022-12-05T01:01:56Z-
dc.date.created2022-12-04-
dc.date.issued2022-07-13-
dc.identifier.citation2022 USENIX Annual Technical Conference, ATC 2022, pp.755 - 771-
dc.identifier.urihttp://hdl.handle.net/10203/301583-
dc.description.abstractWe propose Vigil-KV, a hardware and software co-designed framework that eliminates long-tail latency almost perfectly by introducing strong latency determinism. To make Get latency deterministic, Vigil-KV first enables a predictable latency mode (PLM) interface on a real datacenter-scale NVMe SSD, having knowledge about the nature of the underlying flash technologies. Vigil-KV at the system-level then hides the non-deterministic time window (associated with SSD's internal tasks and/or write services) by internally scheduling the different device states of PLM across multiple physical functions. Vigil-KV further schedules compaction/flush operations and client requests being aware of PLM's restrictions thereby integrating strong latency determinism into LSM KVs. We implement Vigil-KV upon a 1.92TB NVMe SSD prototype and Linux 4.19.91, but other LSM KVs can adopt its concept. We evaluate diverse Facebook and Yahoo scenarios with Vigil-KV, and the results show that Vigil-KV can reduce the tail latency of a baseline KV system by 3.19× while reducing the average latency by 34%, on average.-
dc.languageEnglish-
dc.publisherUSENIX Association-
dc.titleVigil-KV: Hardware-Software Co-Design to Integrate Strong Latency Determinism into Log-Structured Merge Key-Value Stores-
dc.typeConference-
dc.identifier.scopusid2-s2.0-85140958071-
dc.type.rimsCONF-
dc.citation.beginningpage755-
dc.citation.endingpage771-
dc.citation.publicationname2022 USENIX Annual Technical Conference, ATC 2022-
dc.identifier.conferencecountryUS-
dc.identifier.conferencelocationCarlsbad-
dc.contributor.localauthorJung, Myoungsoo-
dc.contributor.nonIdAuthorHwang, Jooyoung-
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0