Destination Physical Machines Research Articles

An energy-efficient load balance strategy based on virtual machine consolidation in cloud environment

In the cloud computing environment, unbalanced utilization of multi-dimensional resources in physical servers generates resource fragmentation, leading to inefficient resource utilization and energy wastage in data centers. The inefficient utilization of resources makes high energy consumption and low quality of service (QoS) in resource management an urgent problem to be solved. We propose a load balancing strategy based on virtual machine consolidation (LBVMC), which aims to reduce the energy consumption and service level agreement (SLA) violation of data centers by balancing the multi-dimensional resource utilization in physical machines (PMs). First, we present a load state classification algorithm for PMs with load abnormality considering current and future loads to reduce unnecessary virtual machine (VM) migrations caused by occasional load fluctuations. Then, we propose a resource-weight based selection model for migratable VMs, which selects appropriate VMs to be migrated based on multi-dimensional resource utilization, and reduces resource fragmentation caused by load imbalance. Finally, we design a VM placement algorithm based on resource fitness and load correlation to deploy VMs on the optimal destination PMs to assure load balancing of the destination PMs after VM placement. We perform simulated experiments under homogeneous, heterogeneous, and bottleneck resource environments. Experimental results show that LBVMC is superior to other strategies in reducing energy consumption and SLA violations and achieves better overall performance.

Raven: Scheduling Virtual Machine Migration During Datacenter Upgrades with Reinforcement Learning

Physical machines in modern datacenters are routinely upgraded due to their maintenance requirements, which involve migrating all the virtual machines they currently host to alternative physical machines. For this kind of datacenter upgrades, it is critical to minimize the time it takes to upgrade all the physical machines in the datacenter, so as to reduce disruptions to cloud services. To minimize the upgrade time, it is essential to carefully schedule the migration of virtual machines on each physical machine during its upgrade, without violating any constraints imposed by virtual machines that are currently running. Rather than resorting to heuristic algorithms as existing work, we propose a new scheduler, Raven, that uses an experience-driven approach with deep reinforcement learning to schedule the virtual machine migration. With our design of the state space, action space and reward function, Raven trains a fully-connected neural network using the cross-entropy method to approximate the policy of choosing a destination physical machine for each virtual machine before its migration. We compare Raven with state-of-the-art algorithms in the literature, and our results show that Raven can effectively shorten the time to complete the datacenter upgrade under different datacenter settings.

Stochastic Load Balancing for Virtual Resource Management in Datacenters

Cloud computing offers a cost-effective and elastic computing paradigm that facilitates large scale data storage and analytics. By deploying virtualization technologies in the datacenter, cloud enables efficient resource management and isolation for various big data applications. Since the hotspots (i.e., overloaded machines) can degrade the performance of these applications, virtual machine migration has been utilized to perform load balancing in the datacenters to eliminate hotspots and guarantee Service Level Agreements (SLAs). However, the previous load balancing schemes make migration decisions based on deterministic resource demand estimation and workload characterization, without considering their stochastic properties. By studying real world traces, we show that the resource demand and workload of virtual machines are highly dynamic and bursty, which can cause these schemes to make inefficient migrations for load balancing. To address this problem, in this paper we propose a stochastic load balancing scheme which aims to provide probabilistic guarantee against the resource overloading with virtual machine migration, while minimizing the total migration overhead. Our scheme effectively addresses the prediction of the distribution of resource demand and the multidimensional resource requirements with stochastic characterization. Moreover, as opposed to the previous works that measure the migration cost without considering the network topology, our scheme explicitly takes into account the distance between the source physical machine and the destination physical machine for a virtual machine migration. The trace-driven experiments show that our scheme outperforms the previous schemes in terms of SLA violation and the migration cost.

A Resource Usage Intensity Aware Load Balancing Method for Virtual Machine Migration in Cloud Datacenters

To provide robust infrastructure as a service (IaaS), clouds currently perform load balancing by migrating virtual machines (VMs) from heavily loaded physical machines (PMs) to lightly loaded PMs. The unique features of clouds pose formidable challenges to achieving effective and efficient load balancing. First, VMs in clouds use different resources (e.g., CPU, bandwidth, memory) to serve a variety of services (e.g., high performance computing, web services, file services), resulting in different overutilized resources in different PMs. Also, the overutilized resources in a PM may vary over time due to the time-varying heterogeneous service requests. Second, there is intensive network communication between VMs. However, previous load balancing methods statically assign equal or predefined weights to different resources, which lead to degraded performance in terms of speed and cost to achieve load balance. Also, they do not strive to minimize the VM communications between PMs. We propose a Resource Intensity Aware Load balancing method (RIAL). For each PM, RIAL dynamically assigns different weights to different resources according to their usage intensity in the PM, which significantly reduces the time and cost to achieve load balance and avoids future load imbalance. It also tries to keep frequently communicating VMs in the same PM to reduce bandwidth cost, and migrates VMs to PMs with minimum VM performance degradation. We also propose an extended version of RIAL with three additional algorithms. First, it optimally determines the weights for considering communication cost and performance degradation due to VM migrations. Second, it has a more strict migration triggering algorithm to avoid unnecessary migrations while still satisfying Service Level Objects (SLOs). Third, it conducts destination PM selection in a decentralized manner to improve scalability. Our extensive trace-driven simulation results and real-world experimental results show the superior performance of RIAL compared to other load balancing methods.

Reinforcement Learning Based APO-PTIRIAL for Load Balancing in Cloud Computing Environment

Power consumption-Traffic aware-Improved Resource Intensity Aware Load balancing (PT-IRIAL) method was proposed to balance load in cloud computing by choosing the migration Virtual Machines (VMs) and the destination Physical Machines (PMs). In this paper, an Artificial Intelligence (AI) technique called Reinforcement Learning (RL) is introduced to find out an optimal time to migrate the selected VM to the selected destination PM. RL enables an agent to find out the most appropriate time for VM migration based on the resource utilization, power consumption, temperature and traffic demand. RL is incorporated into the cloud environment by creating multiple state and action space. The state space is obtained through the computation of resource utilization, power consumption, temperature and traffic of selected VMs. The action space is represented as wait or migrate which is learned through a reward function. Based on the action space, the selected VMs are waiting or migrating to the selected destination PMs.

A Game Based Consolidation Method of Virtual Machines in Cloud Data Centers With Energy and Load Constraints

To reduce energy consumption and balance the resource load of physical machines (PMs) in cloud data centers, we present a game-based consolidation method of virtual machines (VMs) with energy and load (applied load) constraints. First, we test every measured value of the resource load using a t-test to filter outliers. Based on these values, the future resource load is forecast using gray theory. Second, all online PMs are grouped by the number of VMs on them and their future load values. Based on the groupings, a pre-processing algorithm for selecting destination PMs is proposed to determine a set of destination PMs for a VM awaiting migration. Finally, we select the final destination PM for the VM using game-based methods aimed at optimizing overall energy consumption. The experimental results show that our method can reduce energy consumption as well as balance loads without unnecessarily increasing the number of VM migrations.

Distributed Autonomous Virtual Resource Management in Datacenters Using Finite-Markov Decision Process

To provide robust infrastructure as a service, clouds currently perform load balancing by migrating virtual machines (VMs) from heavily loaded physical machines (PMs) to lightly loaded PMs. Previous reactive load balancing algorithms migrate VMs upon the occurrence of load imbalance, while previous proactive load balancing algorithms predict PM overload to conduct VM migration. However, both methods cannot maintain long-term load balance and produce high overhead and delay due to migration VM selection and destination PM selection. To overcome these problems, in this paper, we propose a proactive Markov Decision Process (MDP)-based load balancing algorithm. We handle the challenges of allying MDP in virtual resource management in cloud datacenters, which allows a PM to proactively find an optimal action to transit to a lightly loaded state that will maintain for a longer period of time. We also apply the MDP to determine destination PMs to achieve long-term PM load balance state. Our algorithm reduces the numbers of service level agreement (SLA) violations by long-term load balance maintenance, and also reduces the load balancing overhead (e.g., CPU time and energy) and delay by quickly identifying VMs and destination PMs to migrate. We further propose enhancement methods for higher performance. First, we propose a cloud profit oriented reward system in the MDP model so that when the MDP tries to maximize the rewards for load balance, it concurrently improves the actual profit of the datacenter. Second, we propose a new MDP model, which considers the actions for both migrating a VM out of a PM and migrating a VM into a PM, in order to reduce the overhead and improve the effectiveness of load balancing. Our trace-driven experiments show that our algorithm outperforms both previous reactive and proactive load balancing algorithms in terms of SLA violation, load balancing efficiency, and long-term load balance maintenance. Our experimental results also show the effectiveness of our proposed enhancement methods.

A Survey on Live Migration Techniques of Virtual Machines

Objectives: Cloud computing plays an imperative role in development of technology and makes the data available to the worldwide extent. Virtualization and virtual machines helps in enhancing the utility of the cloud computing to its maximum limit. Method: The method of replication of virtual machine in source physical machine to destination physical machine is known as migration of virtual machines. Live migration is a type of migration that deploys pre-copy, post copy, hybrid approach for transferring load. Finding: Load balancing is the foremost problems in data centres which can be handled by migration. This paper focuses on pre-copy approach and makes an investigation of the various pre-copy methods and its implementation. Applications: It helps in improving the utilization of data centres and load management with minimum downtime and no page faults.