Distributed Parallel To The Research And Implementation Of The Router Load Balancing And Fault-tolerant Technology

With Internet flow growing , the router becomes the blockneck of networks .As the result , improving the router's speed has been one of the emphases on the research. In addition, as the core equipment of Internet , the router's reliability should not been neglected. Based on the above , we designed and implemented the Distributed Parallel Router System(DPRS) ,which is running on Linux and composed of several common personal computers.With the help of distributed architecture , the parallel processing ability of the router is greatly improved and the single Point of failure is resolved. These help the DPRS have high performance ,high reliability, high scalability and low price. In this thesis , we firstly describe the logical structure of DPRS , and then analyze the present load balance technology , what's more ,based on the analysis we introduce the implementation of DPRS ;Secondly , we analyze the fault tolerance technology for distributed system ,including the fault detection technology ,breakpoint setting and two mechanisms about fault restoration , and then we describe the implementation of the special fault tolerance in DPRS. DPRS uses IP as the task assignment granutarlty and the soure IP adrress hashing algorithm to implement the load balance. The fault detection makes use of the heartbeat machnism with checkpoint to detect faults in time ,so the load balance algorithm can reassign tasks in the available nodes.