The web applications have now become one of the most important parts of people’s life. It entered the time-to-market age, that is, the faster released the products, the more the chances that it will increase the company potentialities. System stability test plays an important role of improving the service quality and developing these web applications. There were three well-known strategies to solve this problem: developing application program interface to measure the response time, through open source softwares (e.g. Jmeter), and via the commercial packages (e.g. LoadRunner). LoadRunner is an industry-standard package that has the advantages of graph-based analysis, statistical analysis of the measured data, service-level agreement, and the loading analysis for the client end compared with other strategies. This paper demonstrated that it monitored the service quality of web applications using the smoke and stress tests for travel service applications via LoadRunner. And we proposed a strategy for on-line web server testing. We hope that the practical experience and the information are useful for researchers.

A parallelization of Poisson equation solver is developed by using the parallel algorithm of red-black SOR method performed on GPU by OpenACC. We use the parallel computing solver to solve the fluid-structure interaction problem with direct-forcing immersed boundary method. The speedup for the parallel computing solver is up to 8.62 times. In addition, we present that the optimization of the red-black SOR (RBSOR) algorithm for the problem with the complex-geometry objects is to allocate the same memory on two variables which is to take advantages of accessing data by random indexes and formatted indexes. The implementation of parallel computing RBSOR with OpenACC is efficient to delegate the simulation of fluid-structure interaction problem on the GPU.

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design. Keywords—Artificial Intelligence, machine learning, deep learning, convolutional neural networks

With the development of virtualization technologies, a new type of service named cloud computing service is produced. Cloud users usually encounter the problem of how to use the virtualized platform easily over the web without requiring the plug-in or installation of special software. The object of this paper is to develop a system and a method enabling process interfacing within an automation scenario for accessing remote application by using the web browser. To meet this challenge, we have devised a web-based interface that system has allowed to shift the GUI application from the traditional local environment to the cloud platform, which is stored on the remote virtual machine. We designed the sketch of web interface following the cloud virtualization concept that sought to enable communication and collaboration among users. We describe the design requirements of remote application technology and present implementation details of the web application and its associated components. We conclude that this effort has the potential to provide an elastic and resilience environment for several application services. Users no longer have to burden the system maintenances and reduce the overall cost of software licenses and hardware. Moreover, this remote application service represents the next step to the mobile workplace, and it lets user to use the remote application virtually from anywhere. Keywords—Virtualization technology, virtualized platform, web interface, remote application.

 Abstract — Virtualization technologies are experiencing a renewed interest as a way to improve system reliability, and availability, reduce costs, and provide flexibility. This paper presents the development on leverage existing cloud infrastructure and virtualization tools. We adopted some virtualization technologies which improve portability, manageability and compatibility of applications by encapsulating them from the underlying operating system on which they are executed. Given the development of application virtualization, it allows shifting the user’s applications from the traditional PC environment to the virtualized environment, which is stored on a remote virtual machine rather than locally. This proposed effort has the potential to positively provide an efficient, resilience and elastic environment for online cloud service. Users no longer need to burden the platform maintenances and drastically reduces the overall cost of hardware and software licenses. Moreover, this flexible and web-based application virtualization service represents the next significant step to the mobile workplace, and it lets user executes their applications from virtually anywhere.

 Abstract — Cloud computing is the innovative and leading information technology model for enabling convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort. In this paper, we aim at the development of workflow management system for cloud computing platforms based on our previous research on the dynamic allocation of the cloud computing resources and its workflow process. We took advantage of the HTML5 technology and developed web-based workflow interface. In order to enable the combination of many tasks running on the cloud platform in sequence, we designed a mechanism and developed an execution engine for workflow management on clouds. We also established a prediction model which was integrated with job queuing system to estimate the waiting time and cost of the individual tasks on different computing nodes, therefore helping users achieve maximum performance at lowest payment. This proposed effort has the potential to positively provide an efficient, resilience and elastic environment for cloud computing platform. This development also helps boost user productivity by promoting a flexible workflow interface that lets users design and control their tasks' flow from anywhere.

: Cloud virtualization technologies are becoming more and more prevalent, cloud users usually encounter the problem of how to access to the virtualized remote desktops easily over the web without requiring the installation of special clients. To resolve this issue, we took advantage of the HTML5 technology and developed web-based remote desktop. It permits users to access the terminal which running in our cloud platform from anywhere. We implemented a sketch of web interface following the cloud computing concept that seeks to enable collaboration and communication among users for high performance computing. Given the development of remote desktop virtualization, it allows to shift the user’s desktop from the traditional PC environment to the cloud platform, which is stored on a remote virtual machine rather than locally. This proposed effort has the potential to positively provide an efficient, resilience and elastic environment for online cloud service. This is also made possible by the low administrative costs as well as relatively inexpensive end-user terminals and reduced energy expenses.

Cloud computing is the innovative and leading information technology model for enabling convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort. This paper presents our development on enabling an individual user's desktop in a virtualized environment, which is stored on a remote virtual machine rather than locally. We present the initial work on the integration of virtual desktop and application sharing with virtualization technology. Given the development of remote desktop virtualization, this proposed effort has the potential to positively provide an efficient, resilience and elastic environment for online cloud service. Users no longer need to burden the cost of software licenses and platform maintenances. Moreover, this development also helps boost user productivity by promoting a flexible model that lets users access their desktop environments from virtually anywhere. Keywords—Cloud Computing, Virtualization, Virtual Desktop, Elastic Environment.

The cloud computing technology is developed rapidly in recent years, in which the remote software is delivered as a service and accessed by users using a thin client over the Internet. In order to share the software resources, virtualization technology plays an important role in the cloud computing environment. In this paper our main goal is to provide the basic knowledge about the virtualization technology of desktop and application sharing, and proposed the virtual desktop and application sharing system to provide an efficient, reliable and elastic service platform for cloud computing environment. By developing the virtual desktop and application sharing technology with the cloud cluster platform, it will realize the VDaaS (Virtual Desktop as a Service) and VAaaS (Virtual Application as a Service) for users, and it will also provide more innovative application services for users in the cloud computing environment.

— Security is an interesting and significance issue for popular virtual platforms, such as virtualization cluster and cloud platforms. Virtualization is the powerful technology for cloud computing services, there are a lot of benefits by using virtual machine tools which be called hypervisors, such as it can quickly deploy all kinds of virtual Operating Systems in single platform, able to control all virtual system resources effectively, cost down for system platform deployment, ability of customization, high elasticity and high reliability. However, some important security problems need to take care and resolved in virtual platforms that include terrible viruses, evil programs, illegal operations and intrusion behavior. In this paper, we present useful Intrusion Detection Mechanism (IDM) software that not only can auto to analyze all system’s operations with the accounting journal database, but also is able to monitor the system’s state for virtual platforms.

— The rapid improvement of the microprocessor and network has made it possible for the PC cluster to compete with conventional supercomputers. Lots of high throughput type of applications can be satisfied by using the current desktop PCs, especially for those in PC classrooms, and leave the supercomputers for the demands from large scale high performance parallel computations. This paper presents our development on enabling an automated deployment mechanism for cluster computing to utilize the computing power of PCs such as reside in PC classroom. After well deployment, these PCs can be transformed into a pre-configured cluster computing resource immediately without touching the existing education/training environment installed on these PCs. Thus, the training activities will not be affected by this additional activity to harvest idle computing cycles. The time and manpower required to build and manage a computing platform in geographically distributed PC classrooms also can be reduced by this development.

In recent years there has been renewal of interest in the relation between Green IT and Cloud Computing. The growing use of computers in cloud platform has caused marked energy consumption, putting negative pressure on electricity cost of cloud data center. This paper proposes an effective mechanism to reduce energy utilization in cloud computing environments. We present initial work on the integration of resource and power management that aims at reducing power consumption. Our mechanism relies on recalling virtualization services dynamically according to user’s virtualization request and temporarily shutting down the physical machines after finish in order to conserve energy. Given the estimated energy consumption, this proposed effort has the potential to positively impact power consumption. The results from the experiment concluded that energy indeed can be saved by powering off the idling physical machines in cloud platforms. Keywords—Green IT, Cloud Computing, virtualization, power consumption.

This paper proposes a new approach to offer a private cloud service in HPC clusters. In particular, our approach relies on automatically scheduling users’ customized environment request as a normal job in batch system. After finishing virtualization request jobs, those guest operating systems will dismiss so that compute nodes will be released again for computing. We present initial work on the innovative integration of HPC batch system and virtualization tools that aims at coexistence such that they suffice for meeting the minimizing interference required by a traditional HPC cluster. Given the design of initial infrastructure, the proposed effort has the potential to positively impact on synergy model. The results from the experiment concluded that goal for provisioning customized cluster environment indeed can be fulfilled by using virtual machines, and efficiency can be improved with proper setup and arrangements. Keywords—Cloud Computing, HPC Cluster, Private Cloud, Virtualization

Virtualization Technology is an interesting research topic in current cloud computing and service. Using the Virtualization Technology in cloud or cluster computing can obtain a lot of benefits, such as ability to deploy any virtual platforms rapidly, easiness to manage all precious resources, and cost reduction. In order to discover optimal performance for virtual platforms, several well-known virtual machines are evaluated by standard benchmark tools, including HPC Challenge benchmark and NetPIPE program. In our paper, we will analyze significant experiment results that not only demonstrate the adequacy of virtual machines for High Performance Computing, but also present different performance characteristics for virtualization on cloud environment.

Virtualisation technology is a principal research issue in current cloud computing domain. We can obtain many benefits using the virtualisation technology in cloud and cluster computing, such as the ability to deploy any virtual platforms rapidly, easiness to manage all precious resources, provide customisation services platform and cost reduction. In order to discover optimal computing performance for virtualisation platforms, related virtualisation platforms with several well-known virtual machine tools are evaluated via standard benchmark programmes, including HPC challenge benchmark, NetPIPE and NCHC Application Suite. In this paper, we will analyse and compare significant experiment results that not only demonstrate the adequacy of virtual machines for high-performance computing, but also present different performance characteristics for virtualisation on cloud environment.

Scientific computing has become one of the key players in the advance of modern science and technologies. In the meantime, due to the success of developments in processor fabrication, the computing power of Personal Computer (PC) is not to be ignored as well. Lots of high throughput type of applications can be satisfied by using the current desktop PCs, especially for those in computerized classrooms, and leave the supercomputers for the demands from large scale high performance parallel computations. The goal of this work is to develop an automated mechanism for cluster computing to utilize the computing power such as resides in computerized classroom. The PCs in computerized classroom are usually setup for education and training purpose during the daytime, and shut down at night. After well deployment, these PCs can be transformed into a pre-configured cluster computing resource immediately without touching the existing education/training environment installed on these PCs. Thus, the training activities will not be affected by this additional activity to harvest idle computing cycles. To echo today's energy saving issues, a dynamic power management is also developed to minimize energy cost. This development not only greatly reduces the management efforts and time to build a cluster, but also implies the reduction of the power consumption by such a mechanism.