Research And Implementation For The Key Technologies Of RAM Grid

With the development of computer technologies, great improvements have been achieved in CPU and main memory. The magnetic disk, however, becomes performance bottleneck of the whole computer system because of the relative delay of IO devices. Data-intensive applications with large and random disk access, such as web server and DBMS, require frequently disk accessing, which can decrease the application performance. In order to improve system performance, RAM Grid is proposed to address performance issue in data-intensive applications, which can shares and utilizes huge memory resources in inter-network.In this paper, we research the theory and key technologies in RAM Grid and develop a prototype system. Firstly, an algorithm for self-adaptive control is presented to keep dynamic balance in system. With the algorithm, it classifies nodes by role and makes node load balance. Secondly, a method for modifying function execution flow based on API hooking is presented. With the method, we can hook system call execution flow without modify kernel source. Then, a new method for buffer resources organizing and managing is put forward based on LRU and Hash lists. It speeds up the process of search and replacement in system. Finally, an asynchronous method of data buffer flusing with kernel thread is proposed. It minimizes the performance loss when updating the fresh data in system. Testing results indicate that it can really enhance total performance of the whole system with large random disk accessing.