Exploiting locality in lease-based replicated transactional memory via task migration

Danny Hendler; Alex Naiman; Sebastiano Peluso; Francesco Quaglia; Paolo Romano; Adi Suissa

doi:10.1007/978-3-642-41527-2_9

Exploiting locality in lease-based replicated transactional memory via task migration

Danny Hendler, Alex Naiman, Sebastiano Peluso, Francesco Quaglia, Paolo Romano, Adi Suissa

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

17 Scopus citations

Abstract

We present LILAC-TM, the first locality-aware Distributed Software Transactional Memory (DSTM) implementation. LILAC-TM is a fully decentralized lease-based replicated DSTM. It employs a novel self-optimizing lease circulation scheme based on the idea of dynamically determining whether to migrate transactions to the nodes that own the leases required for their validation, or to demand the acquisition of these leases by the node that originated the transaction. Our experimental evaluation establishes that LILAC-TM provides significant performance gains for distributed workloads exhibiting data locality, while typically incurring little or no overhead for non-data local workloads.

Original language	English
Title of host publication	Distributed Computing - 27th International Symposium, DISC 2013, Proceedings
Pages	121-133
Number of pages	13
DOIs	https://doi.org/10.1007/978-3-642-41527-2_9
State	Published - 1 Dec 2013
Event	27th International Symposium on Distributed Computing, DISC 2013 - Jerusalem, Israel Duration: 14 Oct 2013 → 18 Oct 2013

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	8205 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	27th International Symposium on Distributed Computing, DISC 2013
Country/Territory	Israel
City	Jerusalem
Period	14/10/13 → 18/10/13

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-642-41527-2_9

Cite this

Hendler, D., Naiman, A., Peluso, S., Quaglia, F., Romano, P., & Suissa, A. (2013). Exploiting locality in lease-based replicated transactional memory via task migration. In Distributed Computing - 27th International Symposium, DISC 2013, Proceedings (pp. 121-133). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8205 LNCS). https://doi.org/10.1007/978-3-642-41527-2_9

Hendler, Danny ; Naiman, Alex ; Peluso, Sebastiano et al. / Exploiting locality in lease-based replicated transactional memory via task migration. Distributed Computing - 27th International Symposium, DISC 2013, Proceedings. 2013. pp. 121-133 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c8bd091c0dd54512a9c61a085647a429,

title = "Exploiting locality in lease-based replicated transactional memory via task migration",

abstract = "We present LILAC-TM, the first locality-aware Distributed Software Transactional Memory (DSTM) implementation. LILAC-TM is a fully decentralized lease-based replicated DSTM. It employs a novel self-optimizing lease circulation scheme based on the idea of dynamically determining whether to migrate transactions to the nodes that own the leases required for their validation, or to demand the acquisition of these leases by the node that originated the transaction. Our experimental evaluation establishes that LILAC-TM provides significant performance gains for distributed workloads exhibiting data locality, while typically incurring little or no overhead for non-data local workloads.",

author = "Danny Hendler and Alex Naiman and Sebastiano Peluso and Francesco Quaglia and Paolo Romano and Adi Suissa",

year = "2013",

month = dec,

day = "1",

doi = "10.1007/978-3-642-41527-2_9",

language = "English",

isbn = "9783642415265",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "121--133",

booktitle = "Distributed Computing - 27th International Symposium, DISC 2013, Proceedings",

note = "27th International Symposium on Distributed Computing, DISC 2013 ; Conference date: 14-10-2013 Through 18-10-2013",

}

Hendler, D, Naiman, A, Peluso, S, Quaglia, F, Romano, P & Suissa, A 2013, Exploiting locality in lease-based replicated transactional memory via task migration. in Distributed Computing - 27th International Symposium, DISC 2013, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8205 LNCS, pp. 121-133, 27th International Symposium on Distributed Computing, DISC 2013, Jerusalem, Israel, 14/10/13. https://doi.org/10.1007/978-3-642-41527-2_9

Exploiting locality in lease-based replicated transactional memory via task migration. / Hendler, Danny; Naiman, Alex; Peluso, Sebastiano et al.
Distributed Computing - 27th International Symposium, DISC 2013, Proceedings. 2013. p. 121-133 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8205 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Exploiting locality in lease-based replicated transactional memory via task migration

AU - Hendler, Danny

AU - Naiman, Alex

AU - Peluso, Sebastiano

AU - Quaglia, Francesco

AU - Romano, Paolo

AU - Suissa, Adi

PY - 2013/12/1

Y1 - 2013/12/1

N2 - We present LILAC-TM, the first locality-aware Distributed Software Transactional Memory (DSTM) implementation. LILAC-TM is a fully decentralized lease-based replicated DSTM. It employs a novel self-optimizing lease circulation scheme based on the idea of dynamically determining whether to migrate transactions to the nodes that own the leases required for their validation, or to demand the acquisition of these leases by the node that originated the transaction. Our experimental evaluation establishes that LILAC-TM provides significant performance gains for distributed workloads exhibiting data locality, while typically incurring little or no overhead for non-data local workloads.

AB - We present LILAC-TM, the first locality-aware Distributed Software Transactional Memory (DSTM) implementation. LILAC-TM is a fully decentralized lease-based replicated DSTM. It employs a novel self-optimizing lease circulation scheme based on the idea of dynamically determining whether to migrate transactions to the nodes that own the leases required for their validation, or to demand the acquisition of these leases by the node that originated the transaction. Our experimental evaluation establishes that LILAC-TM provides significant performance gains for distributed workloads exhibiting data locality, while typically incurring little or no overhead for non-data local workloads.

UR - http://www.scopus.com/inward/record.url?scp=84893754869&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-41527-2_9

DO - 10.1007/978-3-642-41527-2_9

M3 - Conference contribution

AN - SCOPUS:84893754869

SN - 9783642415265

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 121

EP - 133

BT - Distributed Computing - 27th International Symposium, DISC 2013, Proceedings

T2 - 27th International Symposium on Distributed Computing, DISC 2013

Y2 - 14 October 2013 through 18 October 2013

ER -

Hendler D, Naiman A, Peluso S, Quaglia F, Romano P, Suissa A. Exploiting locality in lease-based replicated transactional memory via task migration. In Distributed Computing - 27th International Symposium, DISC 2013, Proceedings. 2013. p. 121-133. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-41527-2_9

Exploiting locality in lease-based replicated transactional memory via task migration

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this