Cloud Computing Environment for Engineering and Business Education

  • Published on
    27-Mar-2017

  • View
    213

  • Download
    1

Transcript

Cloud computing environment for engineering and business education Gabriel Raicu1, a, Alexandra Raicu2, b 1Constanta Maritime University, Romania, Department of Economic Engineering in Transports, Mircea cel Batran Street, No. 104, Constanta, 900663, Romania 2Constanta Maritime University, Romania, Department of General Engineering Sciences, Mircea cel Batran Street, No. 104, Constanta, 900663, Romania agabriel.raicu@cmu-edu.eu, balexandra.raicu@cmu-edu.eu Keywords: Cloud computing, simulation environments, engineering, capacity planning, KVM, OpenMX. Abstract. The authors present the development of a scientific cloud computing environment (SCCE) for engineering and business simulations that offers high performance computation capability. The software platform consists of a scalable pool of virtual machines running a UNIX-like (Linux) or UNIX-derivative (FreeBSD) operating systems using specialised software based on modelling engineering processes and focused on business training and predictive analytics using simulations. The use of advanced engineering simulation technology allows engineers to understand and predict the future performance of complex structures and systems designs which can be optimized to reduce risk, improve performance or enhance survivability. A key component of cloud computing in Universities as well as in other research centers: they can share computing resources beyond their technical capabilities. With cloud computing, this allows them all to have access to large scales processing power based on KVM (Kernel based Virtual Machine). Our solution provides a more productive approach: a full scale virtualised computer with scalable storage space and instantly upgradable processing capability. It has more flexibility than other network computing systems and saves precious research time and money. Unlike the existing systems, the scientific community can receive support from a large number of specialists who may contribute by in a collaborative way. Introduction A recent survey carried out by Peer 1 Network Enterprises found that of the 88% of key decision-makers that do not use cloud computing, 39% said it was because they don't know enough about it. So, for those who are still in the dark, what is cloud computing anyway? Cloud computing enables users to work on very powerful virtual comput-ers/store files and software remotely, rather than on a hard drive or server at their office. The fact is many people may already be using cloud compu-ting without realizing it, whether through work or personal use [1]. Some examples of cloud computing applications include software as a ser-vice (SaaS), Customer Relationship Management, File storage, File synchronization and file back-up. It's now possible for businesses to have their own private cloud, which incorporates specific services and is only accessible to specific people. Business simulations allows decision makers to establish a common understanding for the strategic values, change key factors and study how strategies and decisions are made. As an additional value, on the cloud you can test decisions and strategies before they are implemented and to see problems before they actually occur. The hybrid parallelization is also supported which is suitable for virtual cluster consisting of multicore virtual machines placed on several distributed hardware nodes. Each of the cluster nodes can be accessed by remote operators to modify the simulation job or to schedule another scientific experiment. Advanced Materials Research Vol. 837 (2014) pp 651-656Online available since 2013/Nov/08 at www.scientific.net (2014) Trans Tech Publications, Switzerlanddoi:10.4028/www.scientific.net/AMR.837.651All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,www.ttp.net. (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-21/11/14,14:14:47)http://www.scientific.nethttp://www.ttp.netWith cloud computing [2], this allows them all to have access to large scales processing power based on KVM (Kernel based Virtual Machine) [3]. Several of the authors papers address the concept of Virtual Learning Computing Center and distributed applications. Some of the authors projects are related to on-line educational developing activity and simulation for educational purposes. Other authors rely on Amazon Elastic Compute Cloud (EC2) [4] which is a web service that provides resizable compute capacity in the cloud. Our solution provides a more productive approach: a full scale virtualised computer with scalable storage space and instantly upgradable processing capability. It has more flexibility than other network computing systems and saves precious research time and money. The platform was optimised to run simulations software [5]. Not only conventional diagonalization schemes are provided for clusters, but also linear scaling and a low-order scaling methods are supported as value solver. The cluster is also adaptable to other related scientific tasks. Unlike the existing systems, the scientific community can receive support from a large number of specialists who may contribute by in a collaborative way. The benefits of cloud computing As long as they can get on the Internet, staff can access information from home, on the road, from clients offices or even from a smartphone such as a BlackBerry or iPhone. Staff can also work collaboratively on files and documents, even when they are not physically together. Documents can simultaneously be viewed and edited from multiple locations. Cloud computing can be very quick and easy to get up and running. There is no need to buy and install expensive software because it is all ready installed online remotely and you run it from there, not to mention the fact that many cloud computing applications are offered free of charge. The need to pay for extensive disk space is also removed. With cloud computing, you subscribe to the software, rather than buy-ing it outright. This means that you only need to pay for it when you need it, and it also offers flexibility, in that it can be quickly and easily scaled up and down according to demand. A major advantage of using cloud computing (figure 1) for many companies is that because it is online, it offers virtually unlimited storage compared to your servers and hard drive limits. Needing more storage space does not cause issues with your computer upgrades and equipment, usually all you need to do is increase your monthly fee slightly for more data storage. Fig. 1 Cloud Technology Services of cloud computing When we refer about Services of the cloud computing we can talk about Infrastructure, Platform and Software. 652 Modern Technologies in Industrial EngineeringIaaS (Infrastructure as a Service): Offers infrastructure on demand. The infrastructure can be anything from storage servers to applications to operating systems. Buying infrastructure or renting it out in traditional models can be very expensive. When you opt for IaaS, you save a lot on expenses, space, and personnel required to set up and maintain the infrastructure. The cloud service provider takes care of setting up and maintaining the infrastructure. PaaS (Platform as a Service): Offers a platform to clients for different purposes. For example, a platform developers to build, test, and host applications that can be accessed by the end users. The end users may or may not know that the application is hosted on the cloud. As mentioned earlier, the storage space for user data may be increased or decreased per the requirement of the applications. As with the SaaS, you do not need to build the platform. SaaS (Software as a Service): This is the most popular form of cloud services. The service provider offers a software to support the service on offer. The software is built by the service provider while the end users can configure it to suit their needs. The clients (end users) however, cannot change or modify the software. It is basically a backup service that offers software to help people back up their data. Thus, you can use the service without actually having to code or buy the software [6]. Cloud Computing at Constanta Maritime University In the last eight years CMUConstanta Maritime Universityhas invested significant funds in developing a cloud/grid computing for research in fundamental scientific fields and the development of new techniques in naval operations. Cloud features at Constanta Maritime University are (figure 2): Hundreds of TB Core Storage ZFS with dual or triple parity check; Hundreds of GB RAM for Virtual Machines; Dozens of Xeon QCE5430/Xeon 8CX7560 computing cores; Gluster/Lustre/HA redundant data storages; Multiple Distributed nodes, ready for KVM Virtual Machines or OpenVZ Containers; Ability to run every type of Linux/BSD/Windows OS in 32/64 bit envi-ronment; Ability to emulate other CPU architectures. Fig. 2 Cloud at CMU Addressing the needs of extended computing at CMU: Reduced Cost: Cloud technology is paid incrementally, saving organizations money; Increased Storage: Organizations can store more data than on private computer systems; Highly Automated: No longer do IT personnel need to worry about keeping software up to date; Flexibility: Cloud computing offers much more flexibility than past computing methods; Advanced Materials Research Vol. 837 653 More Mobility: Employees can access information wherever they are, rather than having to remain at their desks; Allows IT to Shift Focus: No longer having to worry about constant server updates and other computing issues, government organizations will be free to concentrate on innovation; Green Computing: As cloud computing can always be used to reprovisioning of resources, when you need to expand, you need not buy the infrastructure to increase the carbon emissions by way of using more electricity to cool off the computer resources; Security concerns: Though people doubt cloud computing, clouds tend to be more secure than the traditional business models. Clouds offer realtime backup which results in less data loss. In case of outage, your customers can use the backup servers that sync with the main ones as soon as they are up. Your business gets maximum uptime without any loss of data during the transitions. Other than this, clouds are less prone to hacks and DDoS attacks as people dont know the whereabouts of your data; The Cloud Gives New Value to Old Hardware and Software: You can perform outstanding CPU intensive activity on remote hardware even from ordinary, old desktop or laptop; Faster Deployment: Lower Costs, faster implementation. In the figure 3 we present a CMU cloud desktop interface. Fig. 3 CMU cloud desktop interface Remote session sample on working account: Virtualization type is KVM, Virtual Machine Account for student educational & research activity with the 6 Xeon QCE5430 CPU/6GB RAM. For example, OpenMX (Open source package for Material eXplorer) is a software package for nano-scale material simulations based on density functional theories (DFT), norm-conserving pseudopotentials, and pseudo-atomic localized basis functions. The code is designed for the realization of large-scale ab initio calculations on parallel computers, and thereby we expect that OpenMX can be a useful and powerful tool for nano-scale material sciences in a wide variety of systems such as bio-materials, carbon nanotubes, magnetic materials, and nanoscale conductors [7]. Considerable functionalities are available for calculations of physical properties such as magnetic, dielectric, electric transport properties as listed above. Not only conventional diagonalization schemes are provided for clusters, molecules, slab, and solids, but also linear scaling and a low-order scaling methods are supported as eigenvalue solver. The execution environment is unix and linux, figure 4. All work arrays in the program codes are dynamically allocated with the minimum memory size required by an input file. For large-scale calculations parallel execution by MPI or OpenMX is 654 Modern Technologies in Industrial Engineeringsupported for parallel machines. The hybrid parallelization by OpenMP/MPI is also supported which is suitable for PC cluster consisting of multicore processors. For the execution of OpenMX, you are required to possess pseudo-atomic basis orbitals and pseudopotentials. Fig. 4 CMU cloud desktop applications Conclusion The cloud computing environment can be accessed globally using mobile devices or standard desktops. Researchers can operate the cloud simultaneously as scientific virtual teams and can collaborate in scenario development. The scalable infrastructure allows also simultaneously observation during experiments, figure 5. Unlike the existing systems, the scientific community can receive support from a large number of specialists who may contribute by in a collaborative way. Fig. 5 OpenMX cluster results Advanced Materials Research Vol. 837 655Using the platform in conjunction with OpenMX we can provide a large string of features and capabilities as such: total energy and forces by cluster, band, O(N), and low-order scaling methods, local density approximation (LDA, LSDA) and generalized gradient approximation (GGA) to the exchange-correlation potential, norm-conserving pseudopotentials, variationally optimized pseudo-atomic basis functions, fully and scalar relativistic treatment within pseudopotential scheme, non-collinear DFT, constraint DFT for non-collinear spin and orbital orientation, collinear LDA+U and non-collinear LDA+U methods, macroscopic polarization by Berry's phase, Divide-conquer (DC) method, generalized DC method, and Krylov subspace method for O() eigenvalue solver, simple, RMM-DIIS, GR-Pulay, Kerker, and RMM-DIIS with Kerker's metric charge mixing schemes, exchange coupling parameter, effective screeing medium method, scanning tunneling microscope (STM) simulation, nudged elastic band (NEB) method, full and constrained geometry optimization, electric transport calculation by a non-equilibrium Green's function (NEGF) method, construction of maximally localized wannier functions, NVE ensemble molecular dynamics, NVT ensemble molecular dynamics by a velocity scaling and the Nose-Hoover methods, Mulliken, Voronoi, and ESP fitting analysis of charge and spin densities, analysis of wave functions and electron (spin) densities, dispersion analysis by the band calculation, density of states (DOS) and projected DOS, etc. Scientific cloud computing environment (SCCE) may evolve to future developments on virtual computing platforms. References [1] K. Jorissen, F. D. Vila, J. J. Rehr, A high performance scientific cloud computing environment for materials simulations, Cornell University Library, http://arxiv.org/abs/1110.054. [2] G. Raicu, A. Nita, Systems, methods and tools for knowledge management in virtual universities, E-Comm-Line Conference Proceeding Bucharest, Romania (2008), 175-185. [3] Information on http://www.linux-kvm.org [4] Information on http://aws.amazon.com/ec2 [5] G. Raicu, L. Chiotoroiu. WEB Enabled Simulation, Constanta Maritime University Annals Constanta, Romania, and Volume 8, 163-168 [6] Information on http://www.wikipedia.org [7] Information on http://www.openmx-square.org. 656 Modern Technologies in Industrial EngineeringModern Technologies in Industrial Engineering 10.4028/www.scientific.net/AMR.837 Cloud Computing Environment for Engineering and Business Education 10.4028/www.scientific.net/AMR.837.651 http://dx.doi.org/www.scientific.net/AMR.837http://dx.doi.org/www.scientific.net/AMR.837.651

Recommended

View more >