The Form of High-Performance Computing: A Survey

In this paper, we review and summarize some of the leading computing concepts including cluster, grid, and cloud computing. These ideas relate to bringing computers together to build a more powerful computational engine that is capable of processing multiple tasks concurrently. In cluster computing, a group of computers are located close by and directly linked with extremely high speed connections in order to perform as a single computer system. In grid computing, computer resources from multiple Internet domains are combined to reach a common computational goal. Compared with clusters, grids tend to be more loosely coupled, heterogeneous, and geographically distributed. Cloud computing is an emerging technology allowing a large number of interconnected computers, hardware and/or software to be rented as a commodity like electricity. The service types of cluster, grid, and cloud have been compared. We conclude that cloud computing has the potential to meet various computing demands in a cost effective easy way provided that some crucial issues like privacy, security, migration, and management are properly addressed.

Cloud computing and big data: Technologies and applications

With High-technology and society being developed rapidly, there is a trend which is from digital earth to smart earth. Moreover, smart city is one of most important work in smart earth. Therefore, how to realize or design smart city leaves much to be desired. This paper puts forward some strategies and architectures for designing smart city based on geo-spatial information science and technology (GPS, GIS and RS), IT, communication technology, network technology (smart sensor web and ubiquitous sensor network), spatial data mining, high performance computing (grid computing and cloud computing), GPU, artificial intelligence and pattern recognition. The link and elements of smart city as well as applied key technologies in the future are outlined with some typical application instances, which will meet the versatile requirements of smart city service better.

The emergence of cloud computing has a significant impact on the new technology trends (broadband internet, fast connection and virtualisation) in the world of Information Technology and has become the backbone for the developing network technologies. Its concept is based on high-performance computing with cluster computing, distributed parallel computing and grid computing. Here, we present a comparison among all these techniques along with the application fields and development trends. This will enable us a better thinking about these concepts. In this paper, we aim to pinpoint several challenges and issues in these computing adoption perspectives. Also, we discussed on some of the projects and tools, used to design and development of these models. This study will be helpful to explore the requirements of the users in the computing scenario.

Grid computing is a major research area with strong involvement from both academia and the computing industry. The common vision is that grid computing represents the culmination of truly general distributed computing across various resources in a ubiquitous, open-ended infrastructure to support a wide range of different application areas. Although significant progress has been made in the design and deployment of grids, many challenges still remain before the goal of a user-friendly, efficient, and reliable grid can be realized. Grid research issues cover many areas of computer science to address the fundamental capabilities and services that are required in a heterogeneous environment, such as adaptability, scalability, reliability and security, and to support applications as diverse as ubiquitous local services, enterprise-scale virtual organizations, and internet-scale distributed supercomputing. Cloud computing is also emerging as an alternate platform for large-scale distributed applications where resources are typically provided by a single administrative domain in a pay per-use mode. To some, cloud computing is a natural evolution of grid computing, to others, it is a complementary and perhaps competing technology. Grid and cloud research will greatly benefit from interactions with the many related areas of computer science, making Euro-Par an excellent venue to present results and discuss issues.

At first thought, one may not realize the need for a special issue of Parallel and Distributed Computing Practices focused on Cluster Computing when our community has already invested years of research and spent a wealth of resources on traditional Big Iron supercomputers, for instance, the SGI/Cray Origin and IBM SP. Academia, industry, and national laboratories are still using, and for the foreseeable future, will continue to use, supercomputers to solve both grand-challenge scale and high-throughput applications. Then why the interest and new popularity of research into clusters for high-performance parallel and distributed applications? Clusters incorporate the tremendous successes that are attributable to our parallel and distributed processing community. Over the next decade, clusters will span the entire range of high-performance computing platforms. Instead of being restricted solely to supercomputers for achieving high-performance computing, the spectrum of available platforms will spread out to Beowulf-class clusters, to superclusters, and on to supercomputers. Moving along this continuum represents increasing computational resources, which in turn allows the capability for solving larger classes of applications or instances of problems. Compute-bound applications on workstations often may take advantage of the extra horsepower provided by symmetric multiprocessors. Applications with even larger computational requirements and a high-degree of concurrency may use Beowulf clusters, that is, clusters that are home-built from mass-market, commercial off-the-shelf workstations and networks, for instance, PC's interconnected by Fast Ethernet. A supercluster may contain commodity SMP nodes interconnected with scalable, low-latency, high-bandwidth networks, and is integrated by a system vendor rather than the computational scientist. Applications with greater complexity, for example, grand challenges from the sciences, demand scalable systems with high-performance, interconnects, perhaps coupled with terascale I/O subsystems, and thus, are appropriate for these superclusters. Finally, supercomputers, the largest of the clusters, may provide the ultimate platform for the highest-fidelity models and simulations and data-intensive applications. At each step up this ladder of cluster architectures, the system cost increases approximately by an order of magnitude, but having the high-performance capability to solve the problem of interest may require an entry point at any of these levels, the Beowulf, or supercluster, or even supercomputer platform. Cluster computing with Beowulf- or superclusters, thus, may be one or two orders of magnitude less expensive than a supercomputer, and provide a cost-effective solution and capability that was not available on a workstation. But this cost comparison measures just the basic system price. What about the applications? As recently as a few years ago, the lack of a canonical high-performance architectural paradigm meant that migrating to a new computer system, even from the same vendor, typically required redesigning applications to efficiently use the high-performance system. Thus, the application must be redesigned with parallel algorithms and libraries that are optimized for each high-performance system. This process of porting, redesigning, optimizing, debugging, and analyzing, the application may in fact be prohibitively expensive, and certainly frustrating when the process is completed just in time to greet the next generation system! That was then, this is now. With the successful standardization of system-level interfaces, many of these issues no longer hinder our expedited use of these new, high-performance clusters. The message-passing interface (MPI) is a clear example of a standard that has contributed significantly to the success of cluster computing. Supercomputer vendors working together with the high-performance computing community embarked on a process to standardize the user-level interface and behavior for passing messages between concurrent tasks. Now, using ordinary sequential compilers and vendor-supplied or freely-available implementations of this message-passing interface, applications can be implemented efficiently on one supercomputer and ported with little or no effort to other parallel machines. The message-passing paradigm, thus, provides a reasonably good canonical high-performance model and methodology for implementing high-performance applications. An algorithm that exhibits efficient execution and scaling on one machine then is likely to perform well on another with little or no additional programming effort. Using this cluster computing paradigm, algorithms typically are optimized by minimizing the number of remote memory accesses (reducing the amount of messages passed), and carefully organizing the data placement to increase memory locality. Thus, high-performance algorithms and applications that have been optimized for supercomputers, using the same cluster paradigm, may be appropriate for Beowulf and superclusters. Cluster computing, realizing the importance of a standard high-performance programming environment that applies equally well to both ends of the architectural spectrum, highly leverages these and other successes from parallel and distributed processing research and development. This is in an exciting, new era for high-performance computing where the canonical architectural paradigm stretches from home-built Beowulf clusters, to superclusters, and on to supercomputers. But cluster computing faces some challenges ahead of it. Continued research is required for designing 1) scalable, high-performance algorithms, applications, and methodologies, that can tolerate the communication overheads, and adapt to hybrid, SMP, and heterogeneous, nodes; 2) high-performance computer architectures, scalable interconnection networks, and I/O subsystems; and 3) scalable systems software, compilers, and monitoring tools. Cluster computing has come of age, and it is high time for the parallel and distributed processing community to lay the groundwork for our own legacy of success stories. David A. Bader University of New Mexico

Computing has changed the way humans solve problems. It has changed the world more than any other innovation in the recent past. Over the last couple of years, there has been an increased interest in utilizing computing processor powers to its maximum. Cloud, cluster and grid computing are the most common types of computing. All these share some common features toward achieving utility although marginal differences exist in their processes and approaches. The paper presents the relationship and the differences of grid computing from cloud computing in terms of their architecture, applications, security and tools.

Computer and our life is increasingly dependent on computer technology, our life, learning has changed a lot. As a kind of important science and technology, computer information technology is at an unprecedented pace of development into people's lives step by step and is gradually applied to every field of human society. For nearly a century, with the unprecedented speed, computer technology has been widely applied to industrial production and other fields, in various fields plays an important history. This paper begins with the current situation of computer development, analyzes the reasons of the rapid development of computer technology, and then expounds the development of computer science and technology from two aspects: the intelligent supercomputer, the new high performance computer, and then summarizes the development trend of computer science and technology.

Cloud have wide-ranging implications in several areas of computing, notably massive knowledge, cloud computing has simply up to the highest of any list of technology subjects. Furthermore, it is one of the most important chop-chop developing technologies due to the availability of a wide range of computer infrastructure and services. The expansion of cloud computing has included mainframe computing, cluster and grid computing, distributed and parallel computing technologies, and so on. Cloud computing has mostly supplanted grid, cluster, and mainframe computing. This refers to the delivery of numerous services over the web differently. Such as databases, infrastructure, run time platforms, knowledge storage, servers, networking, and code area unit samples of these resources. Price savings, bigger productivity, speed, efficiency, high performance, high computing power, and security are simply many of the explanations why consumers and businesses embrace the on demand computing. This article goes through the definition of cloud computing as well as its properties, various cloud services, deployment models, and security concerns.

PCI Express는 고속, 저전력 등의 특성으로 인하여 프로세서와 주변 I/O 장치들을 연결하는 업계 표준의 버스 기술이다. PCI Express는 최근 고성능 컴퓨터나 클러스터/클라우드 컴퓨팅 등의 분야에서 시스템 인터커넥션 네트워크로서 그 활용가능성을 검증하고 있는 추세이다. PCI Express가 시스템 인터커넥션 네트워크로서 활용가능하게 된 계기는 PCI Express에 NTB(non-transparent bridge) 기술이 도입되면서부터이다. NTB 기술은 물리적으로 두 PCI Express subsystem을 연결가능하도록 하지만, 필요할 경우 논리적인 격리(isolation)를 제공하는 특징이 있다. 또한, PGAS(partitioned global address space)와 같은 공유 주소 공간(shared address space) 프로그래밍 모델은 최근 멀티코어 프로세서의 보편화로 인하여 병렬컴퓨팅 프레임워크로 각광받고 있다. 따라서, 본 논문에서는 차세대 병렬컴퓨팅 플랫폼을 위하여 PCI Express 환경에서 OpenSHMEM을 구현하기 위한 초기 OpenSHMEM API를 설계 및 구현하였다. 본 연구에서 구현한 15가지 OpenSHMEM API의 정확성을 검증하기 위해서 Github의 openshmem-example 벤치마크의 수행을 통하여 확인하였다. 현재 시중에서는 PCI Express 기반 인터커넥션 네트워크는 가격이 매우 비싸고 아직 일반인이 사용하기 용이하도록 NIC형태로 널리 보급되지 않은 실정이다. 이러한 기술개발 초기단계에서 본 연구는 PCI Express 기반 interconnection network를 RDK(evaluation board) 수준에서 실제로 동작하는 실험환경을 구축하고, 여기에 추가로 최근 각광받는 OpenSHMEM software stack를 자체적으로 구현하였다는 데 의의가 있다. PCI Express is a bus technology that connects the processor and the peripheral I/O devices that widely used as an industry standard because it has the characteristics of high-speed, low power. In addition, PCI Express is system interconnect technology such as Ethernet and Infiniband used in high-performance computing and computer cluster. PGAS(partitioned global address space) programming model is often used to implement the one-sided RDMA(remote direct memory access) from multi-host systems, such as computer clusters. In this paper, we design and implement a OpenSHMEM API based on PCI Express maintaining the existing features of OpenSHMEM to implement RDMA based on PCI Express. We perform experiment with implemented OpenSHMEM API through a matrix multiplication example from system which PCs connected with NTB(non-transparent bridge) technology of PCI Express. The PCI Express interconnection network is currently very expensive and is not yet widely available to the general public. Nevertheless, we actually implemented and evaluated a PCI Express based interconnection network on the RDK evaluation board. In addition, we have implemented the OpenSHMEM software stack, which is of great interest recently.

Arc4nix: A cross-platform geospatial analytical library for cluster and cloud computing

Purpose– Cloud Computing (CC) technology is getting implemented rapidly in the educational sector to improve learning, research and other administrative process. As evident from the literature review, most of these implementations are happening in the western countries such as USA, UK, while the level of implementation of CC in developing countries such as India is rare. Moreover, implementing CC technology in the educational sector require various decisions to be made by the managers of the Information Technology (IT) department such as selecting suitable deployment model, vendor providing cloud service, etc. in their respective university or institute. The purpose of this paper is to attempt to address one such decision. Since, different types of CC deployment are available; selecting a suitable one plays a key role, as it might have an impact on the requirements of various stakeholders such as students, teachers, administrative staff (especially the staff members in the IT department), etc. apart from affecting the overall performance of the facilities such as a laboratory. Naturally, a proper decision by analysing multiple perspectives has to be made while carrying out such strategic initiatives by any educational institute.Design/methodology/approach– A case study methodology has been chosen as the research methodology to discuss and demonstrate the above decision problem that was faced in real time by one of the educational institutes in India, offering high-quality management education. The IT managers of this institute were planning to switch over to CC technology for the computer laboratory and they have to make a decision of choosing suitable alternative CC deployment models such as private cloud (PRC), public cloud (PUC), community cloud (COC), hybrid cloud (HYC), etc. by analysing and comparing them based on various factors and perspectives such as elasticity, availability, scalability, etc. Since, multiple factors are involved in making such a strategic decision, the most commonly used Multi-Criteria Decision Making (MCDM) model – namely, the Analytic Hierarchy Process (AHP) is used as a decision support during the decision making process.Findings– The team of decision makers, who were planning to implement CC in the case institute, found that PRC is best as they believed that it would provide adequate cost savings, apart from providing necessary security to maintain confidential information such as student's detail, grades, etc.Research limitations/implications– The results obtained are based on a single case study. Hence, they cannot be generalized for institutions across educational sector. However, the decision making situation and understanding its impact on the stakeholders of the educational institute can be common across various educational institute.Practical implications– Using a real-life case study of an educational institute, this paper presented a strategic decision making situation, which needs to be considered by the IT managers of the educational institutes when they decide to switch over to CC technology. Various criteria to be considered during the decision making process was identified from the literature review were identified and enumerated. These factors would useful for the IT managers of the different educational institute and they can suitably add or delete these decision criteria as per their requirements and situation at hand. Moreover, the algorithm of AHP, which was used as a decision support, was presented in a step-by-step manner, which should be beneficial for the practitioners to apply the same for similar decision making situations.Originality/value– It is believed that this paper would be the first to report on a strategic decision of choosing the deployment model for CC technology especially in the educational sector. Similarly, this paper would also contribute to the field of CC, as it lists out the decision criteria that are to be considered for making the above decision, which has not got adequate importance. Lastly, this paper is also unique in the realm of AHP because application for a decision problem in the field of CC especially in the educational sector is least reported.

Dew Computing (DC) is a comparatively modern field with a wide range of applications. By examining how technological advances such as fog, edge and Dew computing, and distributed intelligence force us to reconsider traditional Cloud Computing (CC) to serve the Internet of Things. A new dew estimation theory is presented in this article. The revised definition is as follows: DC is a software and hardware cloud-based company. On-premises servers provide autonomy and collaborate with cloud networks. Dew Calculation aims to enhance the capabilities of on-premises and cloud-based applications. These categories can result in the development of new applications. In the world, there has been rapid growth in Information and Communication Technology (ICT), starting with Grid Computing (GC), CC, Fog Computing (FC), and the latest Edge Computing (EC) technology. DC technologies, infrastructure, and applications are described. We'll go through the newest developments in fog networking, QoE, cloud at the edge, platforms, security, and privacy. The dew-cloud architecture is an option concerning the current client-server architecture, where two servers are located at opposite ends. In the absence of an Internet connection, a dew server helps users browse and track their details. Data are primarily stored as a local copy on the dew server that starts the Internet and is synchronized with the cloud master copy. The local dew pages, a local online version of the current website, can be browsed, read, written, or added to the users. Mapping between different Local Dew sites has been made possible using the dew domain name scheme and dew domain redirection.

Cloud computing can be regarded as the integration of grid computing and virtualization technology: namely the use of grid distributed computing processing power of IT resources and build into a resource pool, plus on a mature server virtualization, storage virtualization technology make users can real-time monitoring and deployment of resources. Cloud computing and advanced manufacturing technology is facing many challenges during the process of development in the future, including security and privacy will become a primary issue. This study focused on relations with the important areas of emerging industries, the fast scheduling network resources in the cloud computing and advanced manufacturing technology key issues, as well as cloud computing security management issues related to cloud computing and advanced manufacturing technology.

Survey on Programming Models and Environments for Cluster, Cloud, and Grid Computing that Defends Big Data