| 摘要 |
A server architecture performs concurrent information processing in a server system on a multi-core processor environment. The architecture supports simultaneous processing requests comprising multiple classes of queries and/or executing transactions in an application server and/or database server. Requests, which are made in an asynchronous manner, are structured with hash values to enable similar requests to be grouped together. The similar requests are grouped into a group session. All of the requests in the group session are executed at the same time. Accordingly, similar database inserts can be grouped and executed as a single request. The architecture minimizes thread-switching overhead by exploiting inherent parallelism in the inflowing requests. The threads and requests are de-coupled and hence any lock request only makes the execution threads take up another request instead of waiting until the lock is acquired. As such, the threads never go into sleep/wait mode and system resources are utilized more efficiently. |
