Dynamic Content Allocation for Cloud-assisted Service of Periodic Workloads

• Motivated by improved models for content workloadprediction, in this paper we consider the problem of dynamiccontent allocation for a hybrid content delivery system thatcombines cloud-based storage with low cost dedicated serversthat have limited storage and unmetered upload bandwidth. Weformulate the problem of allocating contents to the dedicatedstorage as a finite horizon dynamic decision problem, and showthat a discrete time decision problem is a good approximation forpiecewise stationary workloads. We provide an exact solution tothe discrete time decision problem in the form of a mixed integerlinear programming problem, propose computationally feasibleapproximations, and give bounds on their approximation ratios.Finally, we evaluate the algorithms using synthetic and measuredtraces from a commercial music on-demand service and giveinsight into their performance as a function of the workload characteristics.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Kommunikationssystem (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Communication Systems (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

Nyckelord

SRA - ICT

SRA - Informations- och kommunikationsteknik

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)