A multi-resource load balancing algorithm for cloud cache systems

Abstract

With the advent of cloud computing model, distributed caches have become the cornerstone for building scalable applications. Popular systems like Facebook [1] or Twitter use Memcached [5], a highly scalable distributed object cache, to speed up applications by avoiding database accesses. Distributed object caches assign objects to cache instances based on a hashing function, and objects are not moved from a cache instance to another unless more instances are added to the cache and objects are redistributed. This may lead to situations where some cache instances are overloaded when some of the objects they store are frequently accessed, while other cache instances are less frequently used.In this paper we propose a multi-resource load balancing algorithm for distributed cache systems. The algorithm aims at balancing both CPU and Memory resources among cache instances by redistributing stored data. Considering the possible conflict of balancing multiple resources at the same time, we give CPU and Memory resources weighted priorities based on the runtime load distributions. A scarcer resource is given a higher weight than a less scarce resource when load balancing. The system imbalance degree is evaluated based on monitoring information, and the utility load of a node, a unit for resource consumption. Besides, since continuous rebalance of the system may affect the QoS of applications utilizing the cache system, our data selection policy ensures that each data migration minimizes the system imbalance degree and hence, the total reconfiguration cost can be minimized. An extensive simulation is conducted to compare our policy with other policies. Our policy shows a significant improvement in time efficiency and decrease in reconfiguration cost.

Proactive VNF Scaling with Heterogeneous Cloud Resources: Fusing Long Short-Term Memory Prediction and Cooperative Allocation

“…In the experiments, we Mathematical Problems in Engineering set the demand quantity of different resource types for each request and gradually increased the number of service sets. From the results compared with [7] (denoted as Single) and [30] (denoted as Priority) we can find that the utilization improvement rate of collaborative resource allocation mode is mostly better. e resource utilization rate of cooperation showed better performance distribution than that of others in the heterogeneous setting resource allocation mode.…”

“…To evaluate the effectiveness of resource utilization, the comparison schemes are [7] and [30], and the former adopts a single resource allocation method, the latter aims at minimizing the standard deviation of all server loads. ey minimize the system imbalance by different resources priority, and the priority is given based on their load distributions.…”

Formally modeling and analyzing cost‐aware job scheduling for cloud data center

“…The work of Adnan and Gupta focuses on comparing the cost of developing and deploying the same application using three different architectures and deployment models with the goal of identifying how different architectures and deployments can affect the infrastructure costs of running and scaling an application in the cloud. The work of Jia et al balances the CPU and memory resources between cache nodes by redistributing the data in the cache nodes, thus minimizing the system reconfiguration overhead and load imbalance. The aforementioned references focus on solving the load balancing problem of cloud data center; the main considerations are response time, overload, user satisfaction, etc.…”

Qualitative Study on the efficiency of Load balancing algorithms in Cloud Environment

“…According to [21], Yu, Ivan, Ricardo, Marta and Dianfu, the paper aims at an algorithm which balances both CPU and memory resources among cache instances by re-distributing stored data. Performance measures: reconfiguration cost, minimize load imbalances Shortcoming: This algorithm works efficiently with homogeneous cache nodes with equal capacities of CPU and Memory…”

“…• In the multi-resource algorithm, the algorithm works efficiently with homogeneous cache nodes with equal capacities of CPU and Memory. So it is necessary to develop an algorithm that works with heterogeneous nodes [21].…”