BIRDS EYE VIEW WEB SERVICES GRID ENABLING TECHNOLOGIES CLUSTER TECHNOLOGY LARGE SCLAE DATA SERVICES

1
(No Transcript)
2
Problem

• Lack of uniform and friendly accessibility from
• any location within secure network
• topologies to utilize the high performance
• computing resources efficiently

3
Solution

• Construct complex scripts which chain
• together multiple applications with slow
  I/O-bound file system interfaces

4
Web Services and Integrated Technologies for
Large Scale Data services

• Presented by
• Satya Priya Patnala

5
About the paper. . .

• Integrating Web Service and Grid Enabling
  Technologies to Provide Desktop Access to
  High-Performance Cluster-Based Components for
  Large-Scale Data Services by Victor P. Holmes,
  Wilbur R. Johnson, David J. Miller at Sandia
  National Laboratories

6
Goal Of The Authors

Web services technology
Grid enabling software
Component framework for parallel computing
Service Oriented Architecture
7
Birds eye view . . .

• Web Services
• Grid Technology
• Data Services

8
Web Services
9
Web Services

• Definition
• A web service is an application or application
  component accessible on the web and intended for
  use by another application a client application

10
Web Services

• HTTP protocol acts as the base for the running of
  web services
• Use XML as a standard language
• They follow a set of standardized rules and
  specifications, making it more portable
• Unique from CORBA, RMI,COM
• Eureka No learning Curve !!!!!

11
Web Services Component Technology

• Web services provide a set of XML-based protocols
  and processes for service discovery (UDDI),
  description (WSDL), and interaction (SOAP).
• In general, we can say
• SOAP UDDI WSDL Web Services

12
Component Technology in Web Services

• SOAP (Simple Object Access Protocol) Sends
  message across different modules
• UDDI (Universal Description, Discovery and
  Integration) is the global lookup for locating
  the services
• WSDL (Web Service Definition Language) method
  through which different services are described

13
Grid Technologies

• According to Gartner A Grid is a collection of
  resources owned by multiple organizations which
  are coordinated to allow them to solve a common
  problem
• To provide a rich set of Data Services, common
  grid environments are beginning to accommodate
  data intensive, application-specific, needs

14
Grid Technologies

• Grid services provides access to distributed
  resources, such as a high-performance computing
  cluster for execution of individual Data
  Services, in a manner that is independent of the
  applications or computing systems being utilized.

15
Grid Technologies

• Any application that wishes to use grid services
  does so by submitting one or more workflow
  definitions to a workflow manager service that
  knows how to
• Locate and construct a request specific to a
  resource.
• Discover what services are available on a
  resource.
• Construct a request for a service on a resource.
• Reliably manage one or more requests once it has
  been submitted.
• Transfer data between resources in a single
  request.

16
Data Services

• Data Services Web services Grid Services
• Transform large data sets to visualize critical
  data areas within large simulation outputs
• Take very large data sets as input and reduce
  them to produce data sets that are smaller

17
Data Services

• Parallel processing services which share data
  objects in memory and minimize disk I/O
• Common thread among users intranet

18
Data Services
19
Component Framework for High Performance Cluster
Based Data Services

• Common Component Architecture
• Services on the cluster are implemented as CCA
  components
• One component Data Service algorithm

20
Ccaffeine Framework

• Common Component Architectures existing
  framework is Ccaffeine
• Ccaffeine consists of -
• Command line controller
• Muxer
• Framework itself

21
Ccaffeine Framework

• Command line controller Accepts commands on
  standard input and passes them to the muxer
• Muxer- provide a mechanism for fanning out the
  commands to each of the compute nodes associated
  with a particular parallel run of the framework
• Framework on the whole

22
WSDL (Web Service Description Languages )

• Uses XML document for describing web services
• Basic Elements
• Data description
• Abstract interface definitions
• Binding to actual implementations

23
Elements In Network Definition

• Types a container for data type definitions
  using some type system (such as XSD).
• Message an abstract, typed definition of the
  data being communicated.
• Operation an abstract description of an action
  supported by the service.
• Port Typean abstract set of operations supported
  by one or more endpoints.
• Binding a concrete protocol and data format
  specification for a particular port type.
• Port a single endpoint defined as a combination
  of a binding and a network address.
• Service a collection of related endpoints

24
Data Service for large scale data manipulation
and visualization
25
End Results

• To provide end users the ability to manage and
  understand simulation results from very large
  complex problems
• Integration of hardware and software components
  to support large scale simulations to analysis
  and ensure safety and reliability of the nations
  nuclear weapon stockpile

26
Future Work

• To populate the suite of Data Services demanded
  by the end users
• To augment the architecture for multiple cluster
  deployment, fault tolerance and multi-site access

27
Thank you