Minimize Cost of Data Transfers Using Bandwidth Reservation on FPVB Paths of Dynamic HPNs

Abstract

Extremely large amounts of data are being generated from various scientific domains because of the wide use of the large-scale applications. After generation, the big datasets usually need to be transferred geographically for collaborative data storage and analysis purposes. After efforts, reserving bandwidth along dedicated links of the high-performance networks (HPNs) has emerged as an efficient way for such extremely large amounts of data transfer. In this paper, we focus on the trade-off between cost and the most common performance parameter, i.e., earliest completion time, of data transfers using bandwidth reservation on fixed paths with variable bandwidth of the dedicated HPNs. We consider the scheduling of two types of bandwidth reservation requests regarding such trade-off: (i) to achieve the minimum data transfer cost given the data transfer deadline, and (ii) to achieve the earliest data transfer completion time given the maximum data transfer cost. We propose two optimal algorithms to optimize the scheduling of individual BRRs of these two types. Extensive simulations are then conducted to compare the overall scheduling performance of the proposed algorithms with two existing algorithms from the perspective of different performance metrics, and the simulation results verify the superior performance of the proposed algorithms.