A Scalable Policy and SNMP Based Network Management Framework

Traditional SNMP-based network management can not deal with the task of managing large-scaled distributed network,while policy-based management is one of the effective solutions in network and distributed systems management.However,cross-vendor hardware compatibility is one of the limitations in policy-based management.Devices existing in current network mostly support SNMP rather than Common Open Policy Service(COPS)protocol.By analyzing traditional network management and policy-based network management,a scalable network management framework is proposed.It is combined with Internet Engineering Task Force(IETF)framework for policy-based management and SNMP-based network management.By interpreting and translating policy decision to SNMP message,policy can be executed in traditional SNMP-based device.

Mechanisms for the Application of Health Services Marketing Policies in Algeria

In this paper, we tried to give health services marketing mix and raise patients＇ interest in health services with the contribution to highlighting the possible application mechanisms of different marketing policies on health services in Algeria. Because we see that, health foundations in Algeria like other foundations live in an environment marked with constantly economic, social, politic, and cultural changes. And to respond to change requirements, especially with the appearance of health services market liberation signs, these foundations are obliged to adapt to ensure their existence and development by giving best services to meet the patients＇ needs in order to gain their satisfaction and loyalty. These ones, who became more conscious about the quality of services they are given quantitatively, qualitatively, and timely.

China＇s Service Economy in the Coming Five Years （2011-2015）

This paper suggests that a new approach should be taken for the development of China＇s service sector. Priorities for this include： appropriate objectives for service sector development, enhanced functionality, integrated development between manufacturing and services, improved regional layout of the service sector, and the emergence of reputable brands and leading enterprises. Well-designed and effective industrial policies, as well as calls for major breakthroughs in institutional mechanisms, must support the development of the service sector. Additionally, excessive burdens and difficult financing channels need to be addressed. More attention should be paid to the construction of the credit system, service standardization, and service statistical accounting as element tasks.

Optimal Design for a Retrial Queueing System with State-Dependent Service Rate

This paper considers a single server retrial queue in which a state-dependent service policy is adopted to control the service rate. Customers arrive in the system according to a Poisson process and the service times and inter-retrial times are all exponentially distributed. If the number of customers in orbit is equal to or less than a certain threshold, the service rate is set in a low value and it also can be switched to a high value once this number exceeds the threshold. The stationary distribution and two performance measures are obtained through the partial generating functions. It is shown that this state-dependent service policy degenerates into a classic retrial queueing system without control policy under some conditions. In order to achieve the social optimal strategies, a new reward-cost function is established and the global numerical solutions, obtained by Canonical Particle Swarm Optimization algorithm, demonstrate that the managers can get more benefits if applying this state-dependent service policy compared with the classic model.

An Optimal Resource Provision Policy in Cloud Computing Based on Customer Profiles

Conventional resource provision algorithms focus on how to maximize resource utilization and meet a fixed constraint of response time which be written in service level agreement（SLA）.Unfortunately,the expected response time is highly variable and it is usually longer than the value of SLA.So,it leads to a poor resource utilization and unnecessary servers migration.We develop a framework for customer-driven dynamic resource allocation in cloud computing.Termed CDSMS（customer-driven service manage system）,and the framework’s contributions are twofold.First,it can reduce the total migration times by adjusting the value of parameters of response time dynamically according to customers’profiles.Second,it can choose a best resource provision algorithm automatically in different scenarios to improve resource utilization.Finally,we perform a serious experiment in a real cloud computing platform.Experimental results show that CDSMS provides a satisfactory solution for the prediction of expected response time and the interval period between two tasks and reduce the total resource usage cost.