Investigating the Schedulability of Periodic Real-Time Tasks in Virtualized Cloud Environment

Abstract

In this paper, we developed a computing architecture and algorithms for supporting soft real-time task scheduling in a cloud computing environment through the dynamic provisioning of virtual machines. The architecture integrated three modified soft real-time task scheduling algorithms, namely, Earliest Deadline First, Earliest Deadline until Zero-Laxity, and Unfair Semi-Greedy. A deadline look-ahead module was incorporated into each of the algorithms to fire deadline exceptions and avoid the missing deadlines, and to maintain the system criticality. The results of the implementation of the proposed algorithms are presented in this paper in terms of the average deadline exceptions, the extra resources consumed by each algorithm in handling deadline exceptions, and the average response time. The results not only suggest the feasibility of the soft real-time scheduling of periodic real-time tasks in cloud computing but that the process can also be scaled up to handle the near-hard real-time task scheduling. INDEX TERMS Real-time, cloud computing, virtual machine, deadline, laxity.

A comparative analysis of resource allocation schemes for real-time services in high-performance computing systems

“…EDF and unfair semi-greedy approach H Alhussian et al 26 proposed cloud computing architecture that consists of two parts: master node (MN) Figure 26. Flow of approximate computation-based RA scheme.…”

“…H Alhussian et al 26 proposed cloud computing architecture that consists of two parts: master node (MN) and VM. In this scenario, it is assumed that MN has VMs pool.…”

Energy-Aware Real-Time Tasks Processing for FPGA-Based Heterogeneous Cloud

“…In such scenario, each SR has a predefined life-time (Lt), within which the request has to be completed. These SRs may be periodic/aperiodic in nature [4]. Periodic SRs appear periodically, after a specific time interval.…”

“…We assume that a server has a maximum power capacity of P W k,max = 150 watts 4 . In idle state, the power consumption of a server is assumed as 75 watts.…”

“…The length of the first time-gap, T G 1 is 45. Using equation 1, we obtain the workload (with min service duration) of each 4. In [31], authors experimentally showed such typical max power consumption SR i (W L i ) for each time-gap as shown in Table 4.…”