Presentation on theme: "+ GamaCloud v1.0 a cloud technology architecture on the box"— Presentation transcript:
+ GamaCloud v1.0 a cloud technology architecture on the box
+ abstract The research aims is develop the cloud technology architecture on the single box. GamaCloud give the examples how to use the re-use/unused hardware and the other computer components by developed a grid architecture technology, which hopes have computation capability in the high performance levels. GamaCloud is developing by design and configure the hardware component. The research consist of three phase, they are configuring on cluster, enhance the performance use grid environment, and establishing the cloud application. The research contributions are to address the green technology issues especially in re-usable technology strategies, and compact design of cloud architecture and system for small medium purposes.
+ Literature The main issues on technology environment today are Cloud technology and in the business are cloud-nomics models [1,3] CERN has established in the year 2005 as a new particle accelerator, the Large Hadron Collider (LHC) They are several Grid Data projects, Globus  and Legion , which were directed on computational grids and distributed data management The Particle Physics Data Grids project has developing a grid infrastructure with capabilities on high-speed data transfers and transparent access . The Grid Physics Network a project objective is pursuing an active program of fundamental IT research to realise data and storage virtualization  Data Grid Environment has implemented in the three layers, there are network project design, workflow management, and data management components .
+ Research Methods Un-used & second hard hardware Cluster ComputerGrid Computer Cloud Technology
+ Milestone #1. Cluster Models The cluster models have involved 30 PC for cluster nodes. In the first experiments, the research use BCCD for the operating system. The BCCD is live CD operating system to run and learn about distributed computing in the cluster modes. The research use BCCD to explore and observe the hardware readiness and configuration strategies. The cluster models has implemented the OpenMosix configuration and shared memory strategies
+ OpenMosix config
+ results The node still use the PC model architectures The cluster models develop by use share memory from the nodes to have the maximum memory capacities. The node still work independently The performance of cluster models limited by the specification of each PC's. The integration system as cluster computing in the first level because of the different of specification in each node. The cluster models need work load and balancing especially on task process.
+ Milestone #2 Grid Models In the second step of the research, we develop from the cluster model into grid models. The research does re-engineering of hardware architecture. The steps focus on configuring the PC based in to the grid environment. We combine the science- forge project  to address the need of data grid environments. The grid model use for combining the software configuration between: Operating system Databases in storage cluster Middle-ware services Massive Parsing Interface applications
+ results NoParametersClusterGrid Values 1Hardware architecturesPC based architecturesCluster, middle- ware, and storage 2Resource collaborationBy default and dedicatedDedicated 4Computation powerNot equal on each nodesManageable Risks 5IntegrationNetwork connectedComputation unit connected 6Network coverage1 networkMulti-tier networks 7ProcessParallelsdistributed
+ Milestone #3 Collecting the PCs Re-assembling + overclocking Placing in the single box
+ GamaBox a cloud technology on the box Simple-medium Hardware Software (bundled on OS) Cloud technology and apps On the single box and access GamaBox Hardware GamaBox Hardware
+ Idea goes to prototypes GamaBox PC Short Cut Icon for UGM Apps & Content Cloud Apps Menus UGOS OS 5.0 Research & community support
+ Publication M Riasetiawan. “GamaCloud: The Development of Cluster and Grid Models based shared-memory and MPI” accepted on CITEE 2012, July 12, 2012, JTETI UGM, Yogyakarta, Indonesia M Riasetiawan. “CloudBox: a cloud technology on the box” 8th e- Indonesia Initiative Forum, April 2012, ITB bandung, Indonesia. M Riasetiawan.”Gamacloud-HPC: A Propose Model of the Architecture of Computational System Based On High Performance and Cluster Computing” accepted on International Symposium on Computational System (ISCS) 2012, ISCS 2012, 15 – 16 May 2012, COECS FMIPA UGM, Yogyakarta
+ Literatures M. Riasetiawan.”Cloud-nomics Indonesia Outlook: Potensi Bisnis berbasis Cloud Services”. The 7 th E-Indonesia Initiative Forum, ITB, Bandung, Indonesia. CERN, The European Organization for Nuclear Research: Golden Bernard,”Cloudnomics: The Economics of Cloud Computing”. accessed at April 29, 2011 I. Foster, C. Kesselman, J. Nick, and S. Tuecke. The Physiology of the Grid: An Open Grid Services Architecture for Distributed Systems Integration WG, Global Grid Forum, June 22, 2002 PPDG : Particle Phyisic Data Grid, GriPhyN : Grid Physics Network,
+ W. Jhonston, D. Cannon, B. Nitzberg. Grids as Production Computing Environments: The Emerging Aspects of NASA’ Information Power Grid. Eight IEEE International Symposium on High Performance Distributed Computing, Redondo M.Riasetiawan, AK. Mahmood.”Managing and Preserving Large Data Volume in Data Grid Environment”.2010 International Conference on Information Retrieval and Knowledge Management (CAMP’10).Shah Alam, Selangor, Malaysia. March 17-18, (IEEE) M. Riasetiawan.”CloudBox: Cloud technology on the box”. The 8 th E- Indonesia Initiative Forum, ITB, Bandung, Indonesia. April 24-25, 2012 M. Riasetiawan, AK. Mahmood.”DALA Project”.The Second International Conference on Distributed Framework and Applications (DfmA) 2010.FMIPA Universitas Gadjah Mada. August 2-3, 2010 (IEEE) M. Riasetiawan, AK. Mahmood.”Science-Forge: a Collaborative Scientific Framework”.IEEE ISIEA.Penang – Malaysia. October 3-6, 2010.