| 摘要 |
A cluster server apparatus is provided capable of continuously carrying out data distribution to terminals even if any cache server is failed, while optimally distributing loads on a plurality of cache servers. A cluster control unit (21) distributes requests from the terminals based on the load on each cache server (10x). The cache server (10x) distributes, to the terminal (4y), requested data (streaming data) if it is stored in a streaming data storage unit (14x), while distributing data from a content server (30) if it is not stored therein. Here, the data distributed from the content server (30) is redundantly stored in the streaming data storage units (14x, 14x+1) of two or more cache servers (10x, 10x+1). One cache server (10x) detects the state of distribution of the other cache server (10x+1, 10x-1) that stores the same data as that stored in the one cache server (10). If the one cache server (10x) becomes unable to carry out distribution, the other cache server (10x+1, 10x-1) continues data distribution instead. <IMAGE> |
