A key challenge in computer networking is how to organize network topology effectively among a large number of servers in the cloud storage system. In a cloud environment, the topology, which is different from the underlying topology, may be established in any form at any potential edge peers. The cloud content delivery network (CDN) always faces problems of complex distributed path creation, cache update, load balancing, etc. To address the problem as a static content delivery, we propose an Improved Heuristic Genetic Algorithm for Static Content Delivery in Cloud Storage (IHGA-SCDCS) based on a resource management model and cost model. The static content delivery in cloud storage is abstracted into mathematical model for set solving problem, which is then solved by an improved Genetic Algorithm (GA). Finally, the optimal solution is reduced to an optimal content delivery program. The simulation experiment, based on CloudSim, shows that IHGA-SCDCS can effectively obtain optimal solution while reducing delivery cost.

Introduction

Reasonable network topology and resource management model can not only improve network performance but also guarantee effectiveness and load balancing for resource allocation, thus improving the performance of cloud storage services. Cloud storage is an Internet service itself, which emphasizes cloud data center providing resource while weakening hardware and software capabilities of terminals [1]. With cloud storage service, all content needed is kept in distance data centers and users access it via network. Thus, the cloud storage service may also use CDN in an accelerated manner, like other Internet services, to achieve higher access efficiency and better user experience [2]. The cloud CDN reduces its own cost using competitive price provided by different cloud. Combined with on-demand service of cloud, the cloud CDN can easily adjust its own storage and bandwidth usage according to the requirement. It may also reduce cost by reducing the quality of service. Many scholars have made preliminary studies on CDN programs based on cloud storage. The most representative one is MetaCDN proposed by Broberg et al.[3], which is a low-cost CDN using a storage cloud resource. The system provides a mechanism to place content on network provided by a different storage cloud service supplier and regularly replies appropriate copy answering requests from users. However, the system does not incorporate a new caching strategy and load balancing algorithm. The F_cache content acceleration cache technique developed by FastWeb just caches and finds optimal match on web acceleration content according to strategy. Within a period of time, it does not access file entity from the source website for repeated access but rather it copies content directly from cache to users, thus effectively improving response and saving bandwidth [4]. The cost of cloud CDN includes bandwidth and storage cost. Nevertheless, existing content delivery methods have not arrived at a reasonable solution to a pricing mechanism of cloud CDN. Meanwhile, the network topology of cloud storage is quite different from that of traditional CDN [5]. To operate cloud CDN, it is necessary to study an efficient content delivery program and reasonable load balancing strategy in cloud storage. The CDN is a whole system made up of four elements: content delivery, load balancing, content management and distributed storage after strategic analysis and implementation. The content delivery strategy is one of the key factors in CDN network planning, whose design directly determines whether the core idea as nearest service of CDN can be realized. In accordance with nearest service principle and edge server load balancing strategy, CDN ensures providing service for resource requests from users with an extremely efficient way.