10 likes | 119 Views
This document explores the integration of middleware in grid computing environments utilizing virtual machines (VMs). It addresses pivotal challenges including image and data management, resource management, and the virtualization of resources, with a focus on seamless access to decentralized storage and support for unmodified applications. Insights into the InVIGO middleware framework highlighted in the 2003 ICDCS proceedings exemplify how VMs can facilitate dynamic instantiation and user transparency, enhancing operational efficiency and resource control across diverse architectures.
E N D
Middleware for Grid Computing On Virtual Machines • Middleware challenges • Virtual Machine = process (VM monitor) + data (VM image and user files) • Image and Data Management • Creation, storage, and transfer of VM images for dynamic instantiation, and user data • Resource Management • Application and resource perspectives • Virtual Networks Computing “In-VIGO” (In Virtual Grid Organizations) • Data management: Grid virtual file systems • Seamless access to decentralized storage services of a grid • User-transparent virtual file systems created on-demand on top of NFS (V2,3) • Supports unmodified binary applications and native O/S clients, servers • Virtualization: • Polymorphism • Manifolding • Multiplexing Virtual resources Physical resources • Resource management • Relational database queries • Resource control via real-time schedules • “Classic” Virtual Machines (VMs) • Many,distinct O/Ss that multiplex a physical resource • E.g. VMware (x86), IBM z/VM (S/390) • Prototype and applications • InVIGO middleware beta (Spring 2003) • Java, Globus, SQL, Apache, VFS • Netcare, nCn, Digital Government, BMI, LSS • Grid computing with VMs1 • Security, isolation • Customization, legacy support • Resource control, site independence Wide-area Testbed Architecture Users: nCn, Netcare, … ‘A’ Virtual back-ends ‘B’ A Service provider ‘S’ B C Y X V1 V2 V3 V4 ‘C’ Internet 64-processor IBM xSeries, VMware Internet (Abilene) Data Server D P1 P2 64-processor IBM xSeries, VMware User ‘X’ Front end ‘F’ Physical server P IBM z800 middleware VM startup Information service data session Image Server I 3.4TB IBM “Shark” 1 “A Case for Grid Computing on Virtual Machines”, R. Figueiredo, P. Dinda, J. Fortes, Proceedings of ICDCS 2003 Northwestern University University of Florida Sponsors Research grants by the National Science Foundation: ANIR NSF Middleware Initiative (NMI), EIA CISE-RR, and by IBM