Cloud and Grids

2,815 views
2,733 views

Published on

Overview of Cloud Computing and its current application to Grid Infrastructures

Published in: Technology
0 Comments
4 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
2,815
On SlideShare
0
From Embeds
0
Number of Embeds
26
Actions
Shares
0
Downloads
184
Comments
0
Likes
4
Embeds 0
No embeds

No notes for slide

Cloud and Grids

  1. 1. e-Ciencia 2009 Valencia, Spain, October 29-30 2009 Grids & Clouds: Perspectives and Early Experiences Ruben S.Montero dsa-research.org Distributed Systems Architecture Research Group Universidad Complutense de Madrid 1/23
  2. 2. Objectives Grids & Clouds: Perspectives and Early Experiences ● Provide an overview of Cloud Computing ● Describe how Clouds can help Grids ● Discuss some experiences using Clouds and Grids 2/23
  3. 3. Cloud Computing in a Nutshell Grids & Clouds: Perspectives and Early Experiences What Who Software as a Service On-demand access End-user to any application (does not care about hw or sw) Platform as a Service Platform for building Developer and delivering web (no managing of the underlying applications hw & swlayers) Infrastructure as a ᄎ Service Delivery of a raw System Administrator computer (complete management of the infrastructure computer infrastructure) Physical Infrastructure 3/23
  4. 4. The Public IaaS Cloud Grids & Clouds: Perspectives and Early Experiences ● Simple Web Interface ● Raw Infrastructure Resources ● Total control of the resources ● Capacity leased in the form of Vms ● Complete Service-HW decoupling ● Pay-as-you-go (On-demand access) ● A single user can not get all the resources ● Multi-tenancy ● Elastic & “infinite” Capacity 4/23
  5. 5. The Public IaaS Cloud Grids & Clouds: Perspectives and Early Experiences Total control of service layout ● Software Stack Service ● Type & Number of components End-Users ● Service Elasticity Virtual Machines Web Server App App App (Load Balancer) Server Server Server Network DBs IaaS Cloud (storage) 5/23
  6. 6. The Private IaaS Cloud Grids & Clouds: Perspectives and Early Experiences A “Public Cloud behind the firewall” ● Security concerns ● Flexible management (consolidation, adaptation, provisioning...) Users The headaches... VM VM VM VM VM VM Virtual Infrastruture Virtual Infrastruture ● Orchestrate: ● Virtualization Cloud API Admin Scheduler ● Networking VIM (e.g. OpenNebula) ● Storage Network ● Admin Interfaces Image Repo ● VM placement (Storage) Physical Infrastructure 6/23
  7. 7. The Hybrid IaaS Cloud Grids & Clouds: Perspectives and Early Experiences • Supplement the capacity of the local infrastructure • Transparent access to the resulting hybrid cloud • Utility Computing dream made a reality! Virtual Infrastruture VM VM VM VM LAN Internet Link (VPN) LAN OpenNebula Public Cloud Physical Infrastructure The Utility supplier 7/23
  8. 8. Current Grid Infrastructures... Grids & Clouds: Perspectives and Early Experiences • High degree of heterogeneity (software & hardware) • High operational costs • Isolate and partition resources contributed to the Grid • Specific environment requirements for different Vos • Users simply do not feel like adopting our execution models (pilot jobs...) Grids are difficult to maintain, operate and use 8/23
  9. 9. Grids, Clouds... and Virtual Machines Grids & Clouds: Perspectives and Early Experiences • A VM is an isolated runtime environment (guest OS and apps) • Hypervisors: Full Virtualized, para-virtualization, HW Virtualization Execution of legacy applications Natural way to deal Applications Applications Applications with the heterogeneity Guest OS Guest OS Guest OS Application / HW decoupling Virtualization (Xen, KVM...) Physical Hardware VMs can be provided with Domains are isolated a Cloud-like model! 9/23
  10. 10. Grids, Clouds... and Virtual Machines Grids & Clouds: Perspectives and Early Experiences • Use VMs as basic building block for Grid Services • Current Trends: • VMs as Job Container • VMs as Grid execution service component • Deal with heterogeneity • Simplify & Improve site management • Give VOs control over the worker-node SW • IaaS interfaces for a Grid Site • Attract business users • Support novel execution models 10/23
  11. 11. Grids, Clouds: VM as a Job Container Grids & Clouds: Perspectives and Early Experiences Features Job request with a VM ● Single LRMS-based environemnt ● Integrated with Grid MW Grid middleware Drawbacks (WMS, CE-CREAM...) ● Jobs and VMs are different ● Need to integrate other LRMS (e.g. LSF, PBS...) resources (network, storage...) Jobs executed in a VM VM ● Do not decouple totally infrastructure from the Grid services ● Can not leverage VM features (e.g. Migration...) Cluster WorkerNodes ● Focused on Job execution 11/23
  12. 12. Grids, Clouds: VM as Grid Service Component Grids & Clouds: Perspectives and Early Experiences Job request VM, network & storage with a VM Management environemnt Load-VO Driven Adaptation Grid middleware (WMS, CE-CREAM...) LRMS (e.g. LSF, PBS...) VIM (OpenNebula) Direct Jobs Public execution Cloud Virtual WN (VO) Cloud WNs External Cloud Physical WNs Physical Infrastructure 12/23
  13. 13. Grids, Clouds: IaaS Interface for a Site Grids & Clouds: Perspectives and Early Experiences Grid User Grid User Industry with a VM Business environemnt Services Pilot job / ssh Grid middleware IaaS Interface LRMS (e.g. LSF, PBS...) VIM (OpenNebula) Virtual WorkerNode Other (web, mail...) Raw machines Physical Infrastructure 13/23
  14. 14. Grids, Clouds: First Experiences Grids & Clouds: Perspectives and Early Experiences VMs as a Job Container • INFN: Workernode on demand "Enabling Distributed Job Submission in Dynamic Virtual Execution Environments for EGEE Users", D.Salomoni, M.Cecchi, A.Ghiselli, A.Italiano, M.Orrù, D.Rebatto, V.Venturi, L.Zangrando VMs as a Grid Service Component • CERN: LSF + VMO/OpenNebula + Custom VM images. Tested with real life Grid experiments (ALICE) "The batch virtualization project at CERN", Sebastien Goasguen, Ewan Roche, Tony Cass and Schwickerath Ulrich. ● NIHKEF: Torque/PBS + OpenNebula + CoW VM repository "Virtual Machines at a Tier-1 site." Sander Klous (Nikhef) IaaS for Scientific Computing • Globus Nimbus – AliEn & STAR experiments http://workspace.globus.org/ 14/23
  15. 15. Grids, Clouds: First Experiences Grids & Clouds: Perspectives and Early Experiences • Study Public Clouds (Amazon EC2) to deploy an EGEE site • EGEE site as a private cloud to deploy Grid services • IaaS Interfaces for EGEE sites • StratusLab will integrate, distribute and maintain a cloud turn-key toolkit for EGEE sites EGEE & RESERVOIR Collaboration • Use Virtual Worker-nodes to provision Grid Services • Explore Hybrid Cloud Computing for Grid Sites • Virtualize a Complete EGEE site 15/23
  16. 16. Resource Provisioning Models Grids & Clouds: Perspectives and Early Experiences Where are the resources provisioned Remote GRID CLOUD from? YOUR PRIVATE Local SITE CLOUD Physical Virtual How are the resources provisioned? 16/23
  17. 17. Conclusions Grids & Clouds: Perspectives and Early Experiences • Virtualization, cloud, and grid are complementary technologies and will coexist and cooperate at different levels of abstraction • Virtualization can solve many obstacles for Grid adoption • Virtualization and cloud do NOT require any modification from the end-user perspectives • Separation between service and infrastructure layers will allow the application of the utility model to Grid computing • The use of Cloud interfaces for Grid sites may attract other users to e-infrastructures • We'll see cloud activities in the future EGI 17/23
  18. 18. Grids and Clouds: Perspectives & Experiences THANK YOU FOR YOUR ATTENTION QUESTIONS? 18/23

×