Sökning: onr:"swepub:oai:DiVA.org:kth-285598" >
Assise: Performance...
Assise: Performance and Availability via Client-local NVM in a Distributed File System
-
- Anderson, Thomas (författare)
- University of Washington
-
- Canini, Marco (författare)
- KAUST
-
- Kim, Jongyul (författare)
- KAIST
-
visa fler...
-
- Kostic, Dejan (författare)
- KTH,Network Systems Laboratory (NS Lab)
-
- Kwon, Youngjin (författare)
- KAIST
-
- Peter, Simon (författare)
- The University of Texas at Austin
-
- Reda, Waleed, 1989- (författare)
- KTH,Network Systems Laboratory (NS Lab)
-
- Schuh, Henry (författare)
- University of Washington
-
- Witchel, Emmett (författare)
- The University of Texas at Austin
-
visa färre...
-
(creator_code:org_t)
- USENIX - The Advanced Computing Systems Association, 2020
- 2020
- Engelska.
- Relaterad länk:
-
https://www.usenix.o...
-
visa fler...
-
https://kth.diva-por... (primary) (Raw object)
-
https://urn.kb.se/re...
-
visa färre...
Abstract
Ämnesord
Stäng
- The adoption of low latency persistent memory modules (PMMs) upends the long-established model of remote storage for distributed file systems. Instead, by colocating computation with PMM storage, we can provide applications with much higher IO performance, sub-second application failover, and strong consistency. To demonstrate this, we built the Assise distributed file system, based on a persistent, replicated coherence protocol that manages client-local PMM as a linearizable and crash-recoverable cache between applications and slower (and possibly remote) storage. Assise maximizes locality for all file IO by carrying out IO on process-local, socket-local, and client-local PMM whenever possible. Assise minimizes coherence overhead by maintaining consistency at IO operation granularity, rather than at fixed block sizes.We compare Assise to Ceph/BlueStore, NFS, and Octopus on a cluster with Intel Optane DC PMMs and SSDs for common cloud applications and benchmarks, such as LevelDB, Postfix, and FileBench. We find that Assise improves write latency up to 22x, throughput up to 56x, fail-over time up to 103x, and scales up to 6x better than its counterparts, while providing stronger consistency semantics.
Ämnesord
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Datorsystem (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Computer Systems (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
Nyckelord
- non-volatile memory
- distributed storage
- file systems
- data centers
- reliability
- fault tolerance
- scalability
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
- Av författaren/redakt...
-
Anderson, Thomas
-
Canini, Marco
-
Kim, Jongyul
-
Kostic, Dejan
-
Kwon, Youngjin
-
Peter, Simon
-
visa fler...
-
Reda, Waleed, 19 ...
-
Schuh, Henry
-
Witchel, Emmett
-
visa färre...
- Om ämnet
-
- TEKNIK OCH TEKNOLOGIER
-
TEKNIK OCH TEKNO ...
-
och Elektroteknik oc ...
-
och Datorsystem
-
- NATURVETENSKAP
-
NATURVETENSKAP
-
och Data och informa ...
-
och Datavetenskap
- Av lärosätet
-
Kungliga Tekniska Högskolan