1 / 52

Grid Computing Middleware

Grid Computing Middleware. Thực hiện: - Tăng Thị Thúy Duyên - Trần Công Đời - Trần Công Thanh. Grid Computing Middleware. Introduction Definitions & functions Toolkit: Globus UNICORE gLite. Grid Computing Middleware. Introduction Definitions & functions Toolkit: Globus UNICORE

susane
Télécharger la présentation

Grid Computing Middleware

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Grid Computing Middleware Thực hiện: - Tăng Thị Thúy Duyên - Trần Công Đời - Trần Công Thanh

  2. Grid Computing Middleware • Introduction • Definitions & functions • Toolkit: • Globus • UNICORE • gLite

  3. Grid Computing Middleware • Introduction • Definitions & functions • Toolkit: • Globus • UNICORE • gLite

  4. Introduction Grid Characteristics: Large-scale Heterogeneity Hardware, OS, network, software environments (languages, libraries, tools...) Complex …

  5. Purposes of Middleware • Open, general-purpose and standard • Standard protocols • Defines the contents and sequence of message exchanges used to request remote operation • Important and essential to achieve the interoperability that Grid depends on • Standard APIs • Interfaces to code libraries • Facilitate construction of Grid components by allowing code components to be reused

  6. Application layer User’s applications

  7. Grid Fabric Protocols Tookits Components  resource (computers…

  8. Sensor Networks and Grid Middleware for Laboratory eScience: science on grid School of Chemistry(University of Southampton)+IBM UK Laboratories

  9. Sensor Networks and Grid Middleware for Laboratory

  10. Introduction

  11. GLORIAD 12

  12. Purposes o Open, general-purpose and standard Standard protocols Defines the contents and sequence of message exchanges used to request remote operation Important and essential to achieve the interoperability that Grid depends on Standard APIs Interfaces to code libraries Facilitate construction of Grid components by allowing code components to be reused

  13. What is Grid middleware? System software between applications and operating system Provide services to application Discovery, storage, execution, information, service integration, resource monitoring, failure detection and recovery,… Hide heterogeneous of the Grid environment Provide standardised interfaces to services.

  14. Grid Middleware Connectivity layer:protocols + services  secure message exchange, authentication,and authorization. Resource layer: protocols  sercure access + monitoring Collective layer: coordination of multiple resources+ defines collections of resources Web services model

  15. Grid Middleware software shared, reused, and extended paradigm  API + tools Challenge : social challenge.

  16. Grid Middleware software

  17. Grid Middle Standardization The Global Grid Forum (GGF)  documentation of “best practices” (technical specifications, user experiences, and implementation guidelines). Open Grid forum (OGF)

  18. Open Grid forum www.ogf.org • Global Grid Forum+ Enterprise Grid Alliance 2006 • 3 meetings are/year:North America, Europe and East Asia • 2010: • OGF28March 15-19 Munich, Germany  • OGF29June 20-22 Chicago, Illinois USA  • OGF30October 25-29 Brussels Belgium 

  19. Standards Produced by OGF GLUE- Grid Laboratory Uniform Environment. SAGA -The Simple API for Grid Application OGSA - Open Grid Service Architecture CDDLM- Configuration Description, Deployment, and Lifecycle Management .....

  20. Grid Computing Middleware • Introduction • Definitions & functions • Toolkit: • Globus • UNICORE • gLite

  21. Globus • Open source software toolkit used to build computational grids and grid-based applications. • Developed by the Globus Alliance • Version • Version 1.0 in 1998 • Version 2.0 in 2002 • … • Version 5.0 in 1/2010

  22. Globus Architecture

  23. Globus - GSI Security Layer • Base on • SSL • PKI(public key infrastructure) • X.509 certificate • Support • Single sign-on • Resource authentication through host certificates • Data encryption • Proxy certificate

  24. Globus - GSI Security Layer

  25. Globus - GSI Security Layer

  26. Globus - Resource Management • Globus Resource Allocation Manager (GRAM) • Functionality • Handles resource allocation, initiates and monitors the lifetime of jobs, reports status of job execution. • Resource management. • Globus Resource Specification Language (RSL): • Minimum memory • Number of CPUs

  27. Globus - Resource Management • Globus Access to Secondary Storage (GASS) • A file-access mechanism • Secure HTTP-based streams • GSI _ enforce access permissions for both data and storage.

  28. Globus - Information services • Properties of the nodes that are connected to the grid • Monitoring and Discovery Service (MDS) • Publishing and querying of resource information • Three tier: • Information providers (IPs) • CPU type, OS version. number of processors, memory available,… • Grid Resource Information Service (GRIS) • Grid Information Index Service (GIIS)

  29. Globus - Data management • Transmitting, storing, and managing massive datasets. • Two component: • GridFTP • Secure, efficient, and reliable data movements in grid environments • GSI support for authenticated data transfer • Replica Location and Management • Multiple locations for the same file throughout the grid • Replica location service (RLS)

  30. Globus

  31. Grid Computing Middleware • Introduction • Definitions & functions • Toolkit: • Globus • UNICORE • gLite

  32. Unicore • A integrated grid computing environment. • Including client and server. • Conceived in 1997 • Functionality • Seamless, secure, and intuitive access to resources • Solid authentication mechanisms • Easy relocation of computer jobs to different platforms

  33. Unicore Architecture

  34. Unicore - Unicore client • Local workstation or PC • Two components: • JPA (job preparation agent) • JMC (job monitor component)

  35. Unicore - Unicore gateway • Single entry point for all UNICORE connections into a Usite • Provides an Internet address and a port

  36. Unicore - Unicore Vsite • Two components: • NJS (network job supervisor) • TSI (target system interface)

  37. Grid Computing Middleware • Introduction • Definitions & functions • Toolkit: • Globus • UNICORE • gLite

  38. gLite • deloyed by The Enabling Grids for E-sciencE project (EGEE) • http://glite.web.cern.ch/glite/ • New version:gLite 3.2 Update 09 has been released on 24.03.2010 • gLite is presently supporting Scientifc Linux 4 and 5, and also Debian 4

  39. gLite (CEOSWGISS-21,Budapest.May 8th 2006)

  40. gLite • new information • Number of Countries = 56Number of Sites = 375 (with Logos = 146)Number of (monitored) Resource Brokers = 124Number of Computing Elements = 601 • In VietNam: • Number of Sites: 2 • Site: VN-IFI-PPS    Longname: VN-IFI-PPS    Latitude / Longitude: 20.0 / 120.0    Computing Element: ce.ifi.vngrid.vinaren.vn • Site: VN-IOIT-HN    Longname: VN-IOIT-HN    Latitude / Longitude: 20.0 / 120.0    Computing Element: ce.ioit.vngrid.vinaren.vn • More information: http://gridportal-ws01.hep.ph.ic.ac.uk/dynamic_information/egee-locations.xml

  41. gLite • combines components developed in various related projects (Condor, Globus, ..). • provides the user with high level services for • scheduling and running computational jobs • accessing and moving data • obtaining information on the Grid infrastructure as well as Grid applications • all embedded into a consistent security framework.

  42. gLite Architecture • A layer between services and resources • gLite follows a Service Oriented Architecture

  43. gLite Architecture

  44. gLite – Access service • All of the gLite services are accessible via APIs and CLIs. • API - Application Programming Interface • CLI - Command Line Interface

  45. gLite - Security Services • Authentication • identify entities (users, systems, and services) • Based on personal X.509 Certificates (issued by trusted CAs). • Authorization • allows or denies access to specific services and resources • three basic authorization models, classified as agent, push and pull. • Auditing • monitoring and providing information for post-mortem analysis of security related events. • Dynamic Connectivity Service • tackle the problems that arise when connectivity is restricted by the resource owner.

  46. gLite - Infomation and Monitoring Services • R-GMA Components: provides a uniform method to access and publish distributed information and monitoring data. • Job Management Services (Application Service) • Java logging service, log4j, Apache/Chainsaw (for other languages). • Discovery Service • Provides a standard set of methods for locating Grid services • Network Performance Monitoring • define a standard interface to monitoring frameworks.

  47. gLite - Job Management Services • Accounting • accumulates information about the usage of Grid resources • Computing element • represent a computing resource (a cluster, supercomputers, single workstations). • Workload Management • comprises a set of Grid middleware components responsible for the distribution and management of tasks across Grid resources • Job Provenance • keep track of the definition of submitted jobs, execution conditions and environment. • Package Manager • is a helper service that automates the process of installing, upgrading configuring, and removing software packages from a shared area

  48. gLite - Data Services • Storage Element (SE): • represent a storage resource • Metadate catalog services: • keep track of the data location and relevant metadata (filesize,…) • data movement services: • allow for efficient managed data transfers between SEs.

  49. Comparison

More Related