Time filter

Source Type

Wu H.-S.,Nanjing University | Wu H.-S.,National Key Laboratory for Novel Technology | Wang C.-J.,National Key Laboratory for Novel Technology | Wang C.-J.,Nanjing University | And 2 more authors.
Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology | Year: 2013

Load balancing is a fundamental problem for cloud computing, multi-processing load balancing session persistence in multi-core environment have drawn more attention and have become a focus. For the issue, based on the idea of Hash Linux kernel network data packets passing, a lock-free multi-processing load balancing architecture is proposed, which avoids the use of locks, and can quickly change the existing single-processing load balancing procedure for multi-processing architecture without structural changes. The theory analysis and experimental results show that the proposed architecture is able to improve the overall performance of load balancing system in multi-core environment. Compared with the traditional shared memory architecture, the proposed is able to get better performance and has stronger applicability.


Wu H.-S.,Nanjing University | Wu H.-S.,National Key Laboratory for Novel Technology | Wang C.-J.,National Key Laboratory for Novel Technology | Wang C.-J.,Nanjing University | And 2 more authors.
Xitong Fangzhen Xuebao / Journal of System Simulation | Year: 2013

Load balancing is the core of virtual resource management and scheduling in cloud computing. To overcome the drawbacks of the existing elastic load balancing in cloud computing, an algorithm of prediction-based elastic load balancing resource management (TeraPELB) was proposed, which not only could dynamically deploy resources more flexibly, but also support load-based trend prediction. Theoretical analysis and simulation experiments show that the required number of virtual machines change in compliance with the change of network load, thus TeraPELB is able to dynamically adjust the processing capacity of back-end server cluster with the applied load, and overcomes the drawback that it might lead network service response to turn slow even no response as hysteresis of the applied virtual machine from the cloud. Compared with the traditional elastic load balancing algorithm, TeraPELB is more reasonable for providing scalability and high availability.

Loading National Key Laboratory for Novel Technology collaborators
Loading National Key Laboratory for Novel Technology collaborators