KiWi: A key-value map for scalable real-time analytics

Dmitry Basin; Edward Bortnikov; Anastasia Braginsky; Guy Golan-Gueta; Eshcar Hillel; Idit Keidar; Moshe Sulamy

doi:10.1145/3399718

KiWi: A key-value map for scalable real-time analytics

Dmitry Basin, Edward Bortnikov, Anastasia Braginsky, Guy Golan-Gueta, Eshcar Hillel, Idit Keidar, Moshe Sulamy

Technion - Israel Institute of Technology, Tel Aviv University

Research output: Contribution to journal › Article › peer-review

Abstract

We present KiWi, the first atomic KV-map to efficiently support simultaneous large scans and real-time access. The key to achieving this is treating scans as first class citizens and organizing the data structure around them. KiWi provides wait-free scans, whereas its put operations are lightweight and lock-free. It optimizes memory management jointly with data structure access. We implement KiWi and compare it to state-of-the-art solutions. Compared to other KV-maps providing atomic scans, KiWi performs either long scans or concurrent puts an order of magnitude faster. Its scans are twice as fast as non-atomic ones implemented via iterators in the Java skiplist.

Original language	English
Article number	16
Journal	ACM Transactions on Parallel Computing
Volume	7
Issue number	3
DOIs	https://doi.org/10.1145/3399718
State	Published - Aug 2020

Keywords

Concurrent data structures
key-value maps

All Science Journal Classification (ASJC) codes

Software
Modelling and Simulation
Hardware and Architecture
Computer Science Applications
Computational Theory and Mathematics

Access to Document

10.1145/3399718

Cite this

@article{a8b1685f85e74c53b242233ddefcd369,

title = "KiWi: A key-value map for scalable real-time analytics",

abstract = "We present KiWi, the first atomic KV-map to efficiently support simultaneous large scans and real-time access. The key to achieving this is treating scans as first class citizens and organizing the data structure around them. KiWi provides wait-free scans, whereas its put operations are lightweight and lock-free. It optimizes memory management jointly with data structure access. We implement KiWi and compare it to state-of-the-art solutions. Compared to other KV-maps providing atomic scans, KiWi performs either long scans or concurrent puts an order of magnitude faster. Its scans are twice as fast as non-atomic ones implemented via iterators in the Java skiplist.",

keywords = "Concurrent data structures, key-value maps",

author = "Dmitry Basin and Edward Bortnikov and Anastasia Braginsky and Guy Golan-Gueta and Eshcar Hillel and Idit Keidar and Moshe Sulamy",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.",

year = "2020",

month = aug,

doi = "10.1145/3399718",

language = "الإنجليزيّة",

volume = "7",

journal = "ACM Transactions on Parallel Computing",

issn = "2329-4949",

publisher = "Association for Computing Machinery (ACM)",

number = "3",

}

TY - JOUR

T1 - KiWi

T2 - A key-value map for scalable real-time analytics

AU - Basin, Dmitry

AU - Bortnikov, Edward

AU - Braginsky, Anastasia

AU - Golan-Gueta, Guy

AU - Hillel, Eshcar

AU - Keidar, Idit

AU - Sulamy, Moshe

PY - 2020/8

Y1 - 2020/8

N2 - We present KiWi, the first atomic KV-map to efficiently support simultaneous large scans and real-time access. The key to achieving this is treating scans as first class citizens and organizing the data structure around them. KiWi provides wait-free scans, whereas its put operations are lightweight and lock-free. It optimizes memory management jointly with data structure access. We implement KiWi and compare it to state-of-the-art solutions. Compared to other KV-maps providing atomic scans, KiWi performs either long scans or concurrent puts an order of magnitude faster. Its scans are twice as fast as non-atomic ones implemented via iterators in the Java skiplist.

AB - We present KiWi, the first atomic KV-map to efficiently support simultaneous large scans and real-time access. The key to achieving this is treating scans as first class citizens and organizing the data structure around them. KiWi provides wait-free scans, whereas its put operations are lightweight and lock-free. It optimizes memory management jointly with data structure access. We implement KiWi and compare it to state-of-the-art solutions. Compared to other KV-maps providing atomic scans, KiWi performs either long scans or concurrent puts an order of magnitude faster. Its scans are twice as fast as non-atomic ones implemented via iterators in the Java skiplist.

KW - Concurrent data structures

KW - key-value maps

UR - http://www.scopus.com/inward/record.url?scp=85092762773&partnerID=8YFLogxK

U2 - 10.1145/3399718

DO - 10.1145/3399718

M3 - مقالة

SN - 2329-4949

VL - 7

JO - ACM Transactions on Parallel Computing

JF - ACM Transactions on Parallel Computing

IS - 3

M1 - 16

ER -

KiWi: A key-value map for scalable real-time analytics

Abstract

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this