Prepartition: Paradigm for the Load Balance of Virtual Machine Allocation in Data Centers

TL;DR

This paper introduces Prepartition, a proactive load balancing paradigm for VM allocation in data centers that improves stability and efficiency by pre-setting process-time bounds and planning migrations in advance.

Contribution

The paper proposes Prepartition, a novel proactive load balancing approach for VM allocation that outperforms reactive methods by reducing instability and achieving better load balance.

Findings

Prepartition reduces process time and migration costs.

Prepartition achieves better load balance than reactive methods.

Experimental results confirm theoretical advantages.

Abstract

It is significant to apply load-balancing strategy to improve the performance and reliability of resource in data centers. One of the challenging scheduling problems in Cloud data centers is to take the allocation and migration of reconfigurable virtual machines (VMs) as well as the integrated features of hosting physical machines (PMs) into consideration. In the reservation model, the workload of data centers has fixed process interval characteristics. In general, load-balance scheduling is NP-hard problem as proved in many open literatures. Traditionally, for offline load balance without migration, one of the best approaches is LPT (Longest Process Time first), which is well known to have approximation ratio 4/3. With virtualization, reactive (post) migration of VMs after allocation is one popular way for load balance and traffic consolidation. However, reactive migration has difficulty to reach predefined load balance objectives, and may cause interruption and instability of service and other associated costs. In view of this, we propose a new paradigm, called Prepartition, it proactively sets process-time bound for each request on each PM and prepares in advance to migrate VMs to achieve the predefined balance goal. Prepartition can reduce process time by preparing VM migration in advance and therefore reduce instability and achieve better load balance as desired. We also apply the Prepartition to online (PrepartitionOn) load balance and compare it with existing online scheduling algorithms. Both theoretical and experimental results are provided.