TY - JOUR
T1 - Upper bounds for multi-level multi-server paging
AU - Dolev, Shlomi
AU - Eyal, Anat
AU - Hendler, Danny
AU - Derbeko, Philip
AU - Kogan-Sadetsky, Marina
N1 - Funding Information:
We thank to the anonymous reviewers for the constructive very detailed deep comments, which improved significantly the results. We also thank EMC and ISF grant number 428/11 for supporting the research.
Publisher Copyright:
© 2018 Elsevier B.V.
PY - 2019/2/1
Y1 - 2019/2/1
N2 - The distributed multi-level multi-server paging problem (DMLMSP) defined in this paper extends and generalizes the classical distributed paging problem [3] to a distributed concurrent multi-level setting, in which multiple servers share caches at multiple levels. The DMLMSP can be used for modeling algorithms for efficient distributed storage systems, in which multiple servers use caches for accelerating access to a centralized storage, maintaining cache coherency across multiple nodes while minimizing access latency and optimizing cache-hit ratio. The DMLMSP model fits basic principles of Non-Uniform Cache Architectures (NUCAs) and can be used for analyzing multiple-level centralized storage. We present an optimal offline algorithm for the DMLMSP model, with minimum number of page faults and minimum makespan, whose time complexity is polynomial in the length of the servers' page request sequences. The new algorithm generalizes and simplifies the state of the art algorithms [3,1].
AB - The distributed multi-level multi-server paging problem (DMLMSP) defined in this paper extends and generalizes the classical distributed paging problem [3] to a distributed concurrent multi-level setting, in which multiple servers share caches at multiple levels. The DMLMSP can be used for modeling algorithms for efficient distributed storage systems, in which multiple servers use caches for accelerating access to a centralized storage, maintaining cache coherency across multiple nodes while minimizing access latency and optimizing cache-hit ratio. The DMLMSP model fits basic principles of Non-Uniform Cache Architectures (NUCAs) and can be used for analyzing multiple-level centralized storage. We present an optimal offline algorithm for the DMLMSP model, with minimum number of page faults and minimum makespan, whose time complexity is polynomial in the length of the servers' page request sequences. The new algorithm generalizes and simplifies the state of the art algorithms [3,1].
KW - Concurrency
KW - Distributed computing
KW - Distributed systems
KW - Multi-level shared cache
KW - Optimal offline polynomial paging algorithm
UR - http://www.scopus.com/inward/record.url?scp=85055753606&partnerID=8YFLogxK
U2 - 10.1016/j.ipl.2018.10.012
DO - 10.1016/j.ipl.2018.10.012
M3 - Article
AN - SCOPUS:85055753606
SN - 0020-0190
VL - 142
SP - 72
EP - 76
JO - Information Processing Letters
JF - Information Processing Letters
ER -