Earticle

Both performance and capacity of the main memory are the key to the computer systems in current architecture. DRAM, which is the most used main memory, can’t extend in capacity for its high energy consumption and repeatedly refresh. Fortunately, some new memory mediums, such as phase-change memory (PRAM), are used to replace traditional DRAM memory. These new memories have many advantages, like low energy consumption, without repeatedly refresh, high density storage, and so on. Therefore, these memories are promising. However, their low read/write performance and limited life are restricted the replacement process. In current time, hybrid memory, which consists of both PRAM and DRAM, is a good choice. In this way, the memory capacity can be extended. So, the most challenge for the hybrid memory is the performance. In this paper, we propose a dynamic hybrid LRU caching scheme (DH-LRU) for the last level cache in PRAM/DRAM hybrid main memory to improve the main memory performance. Compared with traditional cache policies, like LRU, FIFO, RANDOM, CFLRU, our DH-LRU improves performance by 4.6%. Moreover, energy consumption of write and read operation can be reduced up to 88.2%.

In order to study the effect of the internet of things (IOT) on the architecture of logistics service supply chain (LSSC), this paper summarizes the application of IOT in related fields on the basis of the theories of IOT, analyzes the effect of IOT on logistics/service flow, information flow and fund flow in LSSC and the effect on the structure of LSSC, builds the architecture of LSSC based on IOT, and finally forecasts the application prospect of IOT in LSSC.

Improving voltage stability, reducing real power loss (PL) and voltage deviation (VD) are the most important tasks in the operation of electrical power systems. Voltage instability and voltage collapse are the severe problems which may take place because of deficit reactive power at load buses due to increased loading or contingencies. In this paper, the problem of obtaining optimal location and size of SVC is formulated as true multi-objective optimization problem for simultaneous minimization of the two objectives namely real power losses and load bus voltage deviation. The two algorithms real coded genetic algorithm (RCGA) and non-dominated sorting genetic algorithm-II (NSGA-II) with a feature of adoptive crowding distance have been used for solving nonlinear constrained multi-objective optimization problem. Both the algorithms have been used for obtaining optimal location and sizing of SVC. Voltage security of the power system has also been analyzed separately for all placement of SVC to ensure secure operation of the system. The proposed approaches have been implemented on IEEE 30-bus test system. The simulation results of the two algorithms have been compared for solution quality, computational complexity and computational time. It has been found that NSGA-II presents better performance in solving multi-objective optimization problem and also in obtaining a diverse set of solutions which converge near the true Pareto-optimal front. The simulation results of NSGA-II have also been presented to exhibit the capabilities of the algorithm to generate well-distributed Pareto-optimal front.

Cloud Computing is one of the major trends in today’s information technology and the most likely research area. Cloud Computing is a model of sharing resources, software, information. The user access the cloud via internet and all the resources are open to all. Thus, cloud security is now a challenging problem for the cloud service providers. Cryptography is a good solution of the cloud security and several research works are published to ensure secure cloud using cryptographic tools. This paper proposes an authentic communication between a customer and cloud service provider using encryption, decryption and message digests.

The current investment and management mode of power grid construction project is inadequate to the whole optimization, the overall process of the project has obvious time-phased, therefore, the paper strongly advocates to carry out life-cycle management in power grid construction project, and divides the construction project into five stages as preliminary planning, design and planning, construction and operation, scrap and disposal, and evaluate and control the cost of operating stage, investment benefit of decision stage, design stage management and risk of construction stage, and sets up performance evaluation model of power grid project, in order to provide certain practical significance for the establishment of life-cycle management mode of power grid enterprises.

With the rapid growth of internet finance, the credit assessing is becoming more and more important. An effective classification model will help financial institutions gain more profits and reduce the loss of bad debts. In this paper, we propose a new Support Vector Machine (SVM) based ensemble model (SVM-BRS) to address the issue of credit analysis. The model combines random subspace strategy and boosting strategy, which encourages diversity. SVM is considered as a state-of-art model to solve classification problem. Therefore, the proposed model has the potential to generate more accuracy classification. Accordingly, this study compares the ANN, LR, SVM, Bagging SVM, Boosting SVM techniques and experience shows that the new SVM based ensemble model can be used as an alternative method for credit assessing.

In the virtualization based cloud computing environment, multiple computers are allowed to run as virtual machines (VM) in a single physical computer. Efficient scheduling of limited underlying resources has been a key issue. This paper presents a formal method of CPU resources scheduling (FRS). VMs are divided into three resources statuses according to resources requirements and their run information. FRS scheduling is formally scheduling CPU resources according to the resources statuses. The implementation of FRS is confined to the VMM layer, without VM dependency. The evaluation shows that idle CPU resources of VMs are be used by those VMs which need more CPU resources and the CPU resources overall utilization is improved.

With the quickly development of Natural Science Foundation of China (NSFC) foundations, management of mass fund-relevant digital data, such as researchers and institutions, academic achievement, funds and so on, is becoming more and more important for supporting basic scientific research. In this paper, we fully introduce the design and implementation of NSFC cloud platform that will be implemented in the service system of knowledge base. Firstly, the whole cloud platform architecture is presented. Then we introduce the system software and equipment environment of our platform. Lastly, we present the construction of private cloud, network components, storage components, and computing components, respectively, with illustrative figures and tables.

According to the traditional stabilized biconjugate gradient algorithm (BiCGSTAB) deficiency in data locality, this paper proposed a QTBiCGSTAB algorithm whose core idea is that recursively divides sparse matrix with quarter tree into sub-matrix and reorders them, to improve the hit ratio of cache and enhance the algorithm’s efficiency. And the idea is good for algorithm being parallized, that is proved by the numerical experiments later. It mainly shows, firstly, QTBiCGSTAB algorithm is more efficiency than BiCGSTAB, and the speedup would reach 1:330. The target division length would be influnced on the algorithm’s performance; Secondly, for large linear system, parallized QTBiCGSTAB is more efficiency than serial’s.