Advanced topics on cloud computing

With the rapid development of hardware, high-speed network, web programming, distributed and parallel computing, and other technologies, cloud computing has emerged as a technological and commercial reality. In cloud computing, infrastructures, platforms, and software are all services, which can be delivered to users via the cloud. There are still many challenges in cloud computing, including computing models, data centers, security and privacy, virtualization, etc. In this special issue, we accepted 5 papers that addressed various research problems in cloud computing. In the first paper, “How to Achieve Non-Repudiation of Origin with Privacy Protection in Cloud Computing”, Wu et al. studied a security issue in cloud computing, called non-repudiation of origin (NRO) with privacy protection on message originator. They defined two goals of NRO: NRO-I and NRO-II, and showed that existentially unforgeable digital signatures can provide NRO-I but not always NRO-II. This paper also presented a communication protocol accommodating non-repudiation of origin and privacy of message originator. In the second paper, “Developing an Optimized Application Hosting Framework in Clouds”, Shi et al. designed and implemented an application placement framework, called EAPAC, for clouds. The existing application placement algorithms assume that the consumption of system resources is proportional to the workloads, and the authors showed that it may not be true under some circumstances. This EAPAC framework can achieve good performance by allocating application request. The framework can also deal with the issue of resource conflicts among applications when there exist concurrent requests in the system. In the third paper, “A Multi-Objective Ant Colony System Algorithm for Virtual Machine Placement in Cloud Computing”, Gao et al. addressed the issue of virtual machine placement. They presented a multi-objective ant colony system algorithm for the virtual machine placement problem, which can efficiently minimize total resource wastage and power consumption. By performance evaluation, it was shown that the proposed algorithm is more efficient than existing solutions. In the fourth paper, “Reliability in Distributed System”, Ahmed et al. scrutinized various challenges and factors affecting reliability and analyzes research models which synthesize significant solutions to tackle possible factors and various challenges in predicting and measuring reliability of software applications in distributed systems. The authors also summarized fault tolerance and fault/failure forecasting techniques. In the fifth paper, “Applications on Large-Scale Multicore Supercomputers”, Wu and Taylor studied how to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore supercomputers. They presented a performance modeling framework based on memory bandwidth contention time and a parameterized communication model. The validation results for the proposed performance modeling method showed less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore supercomputers.