GRID COMPUTING.ppt

CARREON | GUERRERO | RODRIGUEZ
Grid Computing

2
Grid Computing
 Definition:
 Grid computing is distributed computing
performed transparently across multiple
administrative domains.
 Distributed high-performance computing.
 Large geographically distributed networks of
computers.
 Provides a means for using distributed
resources to solve large problems.
 “What the Web did for communication, grids
endeavor to do for computation.”

3
Grid Computing
 Very general computing applications:
 Database searches and queries.
 Scientific applications.
 Weather prediction.
 Cryptography.
 Business applications.
 Transparency:
 Distributing computational resources among
multiple and widely separated sources and
users is a difficult algorithmic problem.

4
Grid Computing
 Grid Computing is a distributed computing model.
 Grid computing technology integrates servers, storage
systems, and networks distributed within the network to
form an integrated system and provide users with
powerful computing and storage capacity.
 For the grid end users or applications, the grid looks
like a virtual machine with powerful capabilities.
 The essence of grid computing is to manage
heterogeneous and loosely coupled resources in an
efficient way in this distributed system, and to
coordinate these resources through a task scheduler
so they can complete specific cooperative computing
tasks.

5
Grid Computing
 A Grid computing network mainly consists of
these three types of machines:
CONTROL
NODE
PROVIDER
USER

7
Characteristics of Grids
 Grids coordinate resources that are not
subject to centralized control.
 Grids use standard, open, general-purpose
protocols and interfaces.
 Grids deliver high qualities of service.
 http://devresource.hp.com/drc/technical_papers/grid_soa/04.png

8
Grid vs. Parallel Computing
SHARCNet – University of
Western Ontario
compneuro.uwaterloo.ca/beowulf.html
Beowulf cluster

9
Grid vs. Parallel Computing
 Grid computing is
distinguished from parallel
computing on one or more
multiprocessors:
 Parallel computing: locally
“clustered” machines or
large supercomputers.
 Grid computing:
computation across
different administrative
domains.
www.chemistry.msu.edu/Facilities/Supercomputer/

10
Two Tenets of Grid Computing
 Virtualization
 Individual resources, such as computers, disks,
information sources, and applications) are pooled
together and made available by abstractions.
 Overcomes “hard-coded” connections between
providers and consumers of resources.
 Provisioning
 When a request for a resource is made, a specific
resource is identified to fulfill the request.
 The system determines how to meet the need, and
optimizes system performance.

11
Characteristics of Grid Applications
 Data acquired by scientific instruments.
 Data are stored in archives on separate,
perhaps geographically-separated sites.
 Data are managed by teams belonging to
different organizations.
 Large quantities of data (tera- or petabytes)
are collected.
 Software used to analyze and summarize the
raw data.

12
The Importance of
Standardization
 Without standardization, grid computing
practitioners would need to acquire accounts
at many different computer centers, managed
by different organizations.
 Different security and authentication protocols
and accounting practices would have to be
applied.
 Very heterogeneous software environment.

13
Importance of Middleware
 Middleware eases grid users’ experience and
provides them with levels of abstraction.
 Middleware extends the Web’s information
and database management capabilities.
 Allowing remote deployment of computational
resources.

14
Globus Toolkit
 Most widely-used
middleware for grids.
 Open source toolkit for
building computing grids.
 Provides a standard
platform upon which other
services build.
 Provides directory
services, security, and
resource management.
www.globus.org

15
CPU Scavenging
 Unused PC resources worldwide are
harnessed. Also known as shared
computing.
 CPU-scavenging systems gain and lose
machines at unpredictable times as users
interact with their computers, or as network
connections fail.
 CPU-scavengers can migrate jobs to allow
smooth operation.

16
SETI@home
 Search for Extraterrestrial Intelligence
 Goal: to analyze vast amounts of data from
the Arecibo radio telescope.
 Initiated by the Space Sciences Laboratory at
the University of California, Berkeley
www.ras.ucalgary.ca/svlbiimages/arecibo.jpg, www.artscouncil.org.uk/spaceart

17
SETI@home
 Uses a free screen saver,
available to the public.
 When activated, the
screensaver program
downloads time sequences
of radio telescope data and
searches them for radio
sources.
 SETI@home has more than
5 million participants.
 Inspiration for other scientific
applications in need of large
computing resources.

18
SETI@home
 Main purpose: A program downloads and
analyzes radio telescope data.
 Data is recorded at the Arecibo Observatory
in Puerto Rico.
 The data is sent to Berkeley, where it is
processed into units of 107 seconds of data.
 These work units are sent from the
SETI@home server over the Internet to
participating computers around the world for
analysis.

19
SETI@home
 The analysis software can search for signals
with about one-tenth the strength of those
sought in previous surveys, because it makes
use of a very computationally intensive
algorithm.
 Data is merged into a database using
SETI@home computers in Berkeley. Various
pattern-detection algorithms are applied to
search for the most interesting signals.

21
 Berkeley Open Infrastructure for
Network Computing.
 Funded by the National Science
Foundation.
 Used in the SETI project.
 Client-server architecture:
 Client – Used by the computer
supplying resources for one or
more BOINC projects. Performs
the computations.
 Server – System software, such
as database services and
project’s web site.
BOINC

22
Remote Procedure Calls
 Mechanism by which the server
communicates with the client in BOINC.
 Like a regular function call or method
invocation, but one computer executes the
function on another computer.

23
Remote Procedure Calls - Examples
 Return screensaver mode:
get_screensaver_mode(int& status)
 Get a list of results for jobs in progress:
get_results(RESULTS&)
 Get a list of file transfers in progress:
get_file_transfers(FILE_TRANSFERS&)
 Get the client’s current state:
get_state(CC_STATE&)

24
Human Proteome Folding Project
(HPFP)
 Goal: to predict the structure of human proteins.
 Devised at the Institute for Systems Biology,
University of Washington.
 Produces the likely structures for each of the
proteins using a set of predefined rules.
 Improved knowledge of human proteins is
important in developing new therapies.
 Officially completed on July 18, 2006.
 Second stage now underway.

25
Human Proteome Folding Project
http://msnbcmedia.msn.com/j/msnbc/Components/Photos/041116/folding2.hmedium.jpg
http://ndg.gunzclan.org/Charlotte/graphics/2/images/IMG0153_JPG.jpg
Typical desktop screensaver
setup for HPFP
WCG desktop console - users monitor
progress on protein-folding project.

26
Business Applications
 Business application grid (BAG).
 Major focus is using existing grid computing
technologies to unite all of an organizations
desktops, workstations, servers, printers,
peripherals, etc., to perform useful work
during idle time.
 Usually focused on well-defined problems:
 Calculating performance averages for a
mutual fund.
 Reducing processing time in wealth
management systems.
 Database applications.

27
Business Applications
 A large financial services company uses
specialized grid software for new corporate
banking applications.
 Oracle Corporation offers a grid database
system.

28
Business Grid Middleware
 Provides an IT-level infrastructure to support
business applications.
 Middleware provides services for composing,
submitting, and managing business
applications.
 Business functions (e.g. credit card
authorization and shipping-and-handling
services) are not provided.
 Globus Toolkit 4 makes it easier to build an
application that taps into existing distributed
computing resources (e.g. servers, storages,
databases).

29
Conclusions
 Grid computing is an “enabling technology”
that is rapidly gaining popularity in:
 Science.
 Medicine.
 Engineering.
 Business and financial applications.
 Many software vendors offer grid computing
toolkits and middleware.
 In 2004, 20% of companies were seeking grid
computing solutions (Evans Data Corp.).

30
Benefits of Grid Computing
 Collaboration.
 Increased productivity.
 Efficient use of resources and storage.
 Cost-effectiveness.
 Heterogeneous environments.
 Failure tolerance.
 Transparency.

31
ADVANTAGES OF GRID
COMPUTING
 It is not centralized, as there are no servers
required, except the control node which is just
used for controlling and not for processing.
 Multiple heterogeneous machines i.e.
machines with different Operating Systems
can use a single grid computing network.
 Tasks can be performed parallelly across
various physical locations and the users don’t
have to pay for them (with money).

32
Challenges
 Lack of control over resources, administration.
 Security.
 Middleware.
 Network failures.
 Cultural issues.

33
DISADVANTAGES OF GRID
COMPUTING
 The software of the grid is still in the involution
stage.
 A super fast interconnect between computer
resources is the need of hour.
 Licensing across many servers may make it
prohibitive for some applications.
 Many groups are reluctant with sharing
resources .

34
Open grid services architecture
 OGSA – standard for grid-based applications.
 Framework for meeting grid requirements.
Application specific grid services
web services
application specific
OGSI services: naming, service data (metadata)
OGSA services: directory, management, security
service creation and deletion, fault model, service groups GridService
e.g.
interfaces
e.g. astronomy, biomedical informatics, high-energy physics
Factory
grid service interfaces
standard
Open-grid services infrastructure

35
Globus toolkit
http://gdp.globus.org/gt3-tutorial/multiplehtml/ch01s04.html
Restricts access to grid services so that
only authorized clients can use them.
Provides another layer of security on top
of firewalls.
Job management – checking status of
jobs, pausing, stopping if necessary.
Index services – helping to locate grid
resources to meet specific needs.
Reliable file transfer service (RFT) –
performs large file transfers from a client
to a grid service.
Replica management – keeps track of
subsets of large data sets that are
being worked on.
Other non-GT3 services can run on
top of the GT3 architecture.
Low-level functions

36
Other grid tools
 Resource management:
 Grid Resource Allocation and Management
Protocol (GRAM)
 Information Services:
 Monitoring and Discovery Service (MDS)
 Security Services:
 Grid Security Infrastructure (GSI)
 Data Movement and Management:
 Global Access to Secondary Storage (GASS)
and GridFTP

37
World-Wide Telescope (2002)
 Goal: deployment of data resources shared
by astronomers.
 Data:
 Archives of observations over a particular
period of time, part of the EM spectrum, and
area of the sky.
 Observations collected at different sites
around the world.
 Data on same celestial objects are combined
over different periods of time and different
parts of the EM spectrum.

38
World-Wide Telescope (2)
 Data archives ( terabyte) managed locally by
the teams that collect the data.
 As data is acquired, it is analyzed and stored
as transformed data so that it can be used by
remote astronomy sites.
 Librarian role of scientists.
 Metatdata is required to describe:
 Time the data was collected.
 Part of the sky observed.
 Instruments used.

39
WCG ongoing projects
 FightAIDS@Home
 Launched by WCG in 2005.
 Each computer processes one potential drug molecule
and tests how well it would dock with HIV protease,
inhibiting viral reproduction.
 Human Proteome Folding Phase 2
 Released in 2006.
 Extension of HPF1, focusing on human-secreted
proteins.
 Better protein models, but more computationally
intensive.

40
World Community Grid (WCG)
 Goal: to create the world's largest public computing
grid for humanitarian concerns.
 Administered and funded by IBM.
 Platforms: Windows, Linux, and Mac OS X.
 Uses the idle time of Internet-connected desktop
computers.
 The agent works as a screen saver (like
SETI@home), only using a computer's resources
when it would otherwise be idle, and returning
resources to the users when requested.
 Projects are approved by an advisory board:
representatives of major research institutions,
universities, UN, WHO.

41
WCG – Smallpox research
 Completed project.
 WCG largely began due to the success of this
project in shaving years off research time.
 Analysis of therapeutic candidates to fight the
small virus.
 About 35 million potential drug molecules were
screened against several smallpox proteins,
resulting in 44 new potential treatments.

42
WCG Ongoing projects
 Help Defeat Cancer (2006)
 Processes large numbers of tissue samples
using tissue microarrays.
 Genome Comparison (2006)
 Compares gene sequences of different
organisms to find similarities.
 Goal: determining the purpose of specific gene
sequences in particular functions by
comparing it with similar sequences with
known functions in another organism.

43
Other grid projects
Description of the project Reference
1. Aircraft engine maintenance using fault histories and
sensors for predictive diagnostics
www.cs.york.ac.uk/dame
2. Telepresence for predicting the effects of
earthquakes on buildings, using simulations and test sites
www.neesgrid.org
3. Bio-medical informatics network providing
researchers with access to experiments and visualizations of results
nbcr.sdsc.edu
4. Analysis of data from the CMS high energy particle
detector at CERN by physicists world-wide over 15 years
www.uscms.org
5. Testing the effects of candidate drug molecules for
their effect on the activity of a protein, by performing parallel
computations using idle desktop computers
[Taufer et al. 2003]
[Chien 2004]
6. Use of the Sun Grid Engine to enhance aerial
photographs by using spare capacity on a cluster of web servers
www.globexplorer.com
7. The butterfly Grid supports multiplayer games for
very large numbers of players on the internet over the Globus toolkit
www.butterfly.net
8. The Access Grid supports the needs of small group
collaboration, for example by providing shared workspaces
www.accessgrid.org

44
Requirements of grid systems
 Remote access to resources, specifically, to
archived data.
 Data processing at the site where the data is
managed.
 Remote requests (queries) result in a
visualization or results from a small quantity of
data.
 Resource manager of a data archive create
instances of services when they are needed.
 Similar to distributed object model, where
servant objects are created when needed.

45
Requirements of grid systems (2)
 Metadata to describe characteristics of
archived data.
 Directory services based on the metadata.
 Software for:
 Query management.
 Data transfer.
 Resource reservation.

GRID COMPUTING.ppt

More Related Content

What's hot (20)

Similar to GRID COMPUTING.ppt (20)

Recently uploaded (20)

GRID COMPUTING.ppt