摘要
基于单位时间开销代价矢量,提出同时考虑地形数据请求在服务器队列中的等待时间与服务器对请求的处理时间的最小总代价分布式算法。算法计算转发请求给每台服务器的概率空间,并根据地形数据请求到达时临时生成的随机数在概率空间中的落点确定转发请求的目标服务器。通过配置单位时间开销代价矢量,可以灵活地维护与升级集群服务器,具有良好的可扩展性。仿真结果表明最小总代价算法能在大规模、高强度的地形漫游中均衡地分发请求,使集群服务器充分发挥其优势从而获得最小的漫游响应时间。仿真结果还表明在低强度地形漫游时,各种负载均衡算法表现大致相同;在高强度地形漫游时,设计负载均衡算法最好考虑服务器端队列的排队情况。
A distributed least cost algorithm is proposed by taking into account the waiting time of terrain tile request in server queue and the process time of it in server, based on an unit time cost vector. Dispatch probability space for each server is calculated and target server is determined through the landing point of a temporarily gener- ated random number in the probability space. The scalability of the algorithm is satisfactory and the maintenance and update of clustered servers is flexible by configuring the unit time cost vector. Extensive simulations indicate that, for terrain tile requests of large-scale and high intensity, the least cost algorithm can balance servers' load efficiently and response the requests with minimal time. Simulation results also show that under low intensity requests, the considered load balancing algorithms behave approximately the same, while under high intensity request, it is better to design load balancing by considering queue situation of clustered servers
出处
《测绘学报》
EI
CSCD
北大核心
2009年第3期242-249,共8页
Acta Geodaetica et Cartographica Sinica
基金
国家973计划(2004CB318206)
国家863计划(2007AA12Z214)