Title: Internet%202%20and%20the%20Grid
1Internet 2 and the Grid
- The Future of Computing for
- Big Science at UNM
Timothy L. Thomas UNM Dept of Physics and
Astronomy
2Acknowledgements Thanks to
- Susan Atlas
- Bob Ballance
- Chris Jordan
- Terry Loring
- Brian Smith
- Art St. George
- Lou Sullo
- Maggie Werner-Washburne
- Susan Wilson
- Terry Yates
- for useful discussions.
3Outline Deconstructing the Title
- Internet 2
- The Grid
- Computing
- Big Science
- at UNM
- The Future
4Grokking The Grid
- Grok v. To perceive a subject so deeply that
one no longer knows it, but rather understands it
on a fundamental level. Coined by Robert
Heinlein in his 1961 novel, Stranger in a Strange
Land.
(Quotes from a colleague of mine) Feb 2002
This grid stuff is garbage. Dec 2002 Hey,
these grid visionaries are serious!
5So what is a Grid?
6(No Transcript)
7(No Transcript)
8(No Transcript)
9- Ensemble of distributed resources acting
- together to solve a problem
- The Grid is about collaboration, about
- people working together.
- Linking people, computing resources, and sensors
/ instruments - Idea is decades old, but enabling technologies
are recent. - Capacity distributed throughout an
infrastructure - Aspects of Grid computing
- Pervasive
- Consistent
- Dependable
- Inexpensive
10- History of Grids a topic of its own
- Corp. for Natl. Research Initiatives Study
Development of societal structures coupled to - the development of infrastructure
- Railroads
- Telegraphs and telephones
- Electric power
- Banking
- The Internet
- Dynamic the emergence of pooling and
caching drives the evolution of middleware. - Why Chicago?
11- Virtual Organizations (VOs)
- Security implications
- Ian Fosters Three Requirements
- VOs that span multiple administrative domains
- Participant services based on open standards
- Delivery of serious Quality of Service
12- Open Standards
- The technical basis for The Grid.
- Its all about standards. CJ
- Global Grid Forum a open standards body
- OGSA / OGSI
- Web services and WSDL
- Globus toolkit version 3 hopes to be
- analogous to TCP/IP.
13- Quality of Service (QoS)
- A Grid should exceed the sum of its parts,
- not be bottlenecked by one of them.
- The Dream Capability, as well as Capacity
- The Grid as a way to solve Really Hard Probems
- Marc Ingbur problems you shouldnt be able to
do - Foster/Kesselman theres 105 out there
- Capability computing is a form of QoS !
14Grid Metaphores
- A Grid allows infrastructures and computations
that can adapt and evolve. - As opposed to a just a "Cyber Infrastructure, I
think more in terms of a "Cyber Ecosystem ! - In any case, new way of organizing human
enterprise - in the 21st centuryboth commercial and
scientific.
15Kinds of Grids
- The Grid will not emerge all at once it
- will grow and evolve like the internet did
- It will form as an intergrid
16Kinds of Grids
- Foster and Kesselman identify several categories
of Grids, - by way of examples
- Government a National Grid
- relatively small numbers of users
- strategic computing reserves
- national collaboratory
- An HMO
- A private Grid
- Life-critical applications
- Specialized instrumentation e.g., CAT scanners
- Privacy issues
17- Material science collaboratory
- Smaller-scale, open scientific research
- Very dynamic membership
- Diverse specialized instrumentation
- Synchrotron light sources
- Advanced microscopes
- Computational market economy
- An open public Grid
- Very large, very dynamic
- More loosely-couple computational resources
18Examples existing Grids
- ( maybe see Hype)
- SETI_at_home
- FightAIDS_at_Home
- Folding_at_Home
- Genome_at_Home
- United Devices Corp w/ Intel Oxford - cancer
research project - U. of Wisc.-centered Condor Pools
- NASA Information Power Grid
- National Earthquake Engineering Simulations
(NEES) Grid
19High Energy Physics Grids
- GriPhyN (NSF)
- CS research focusing on virtual data, request
planning - Virtual Data Toolkit Delivery vehicle for
GriPhyN products - iVDGL International Virtual Data Grid
Laboratory (NSF) - A testbed for large-scale deployment and
validation - Particle Physics Data Grid (DOE)
- Grid-enabling six High-Energy/Nuclear Physics
experiments - EU Data Grid (EDG) Applications areas
- Particle physics
- Earth and planetary sciences "Earth
Observation - Biology
- GLUE Grid Laboratory Uniform Environment
- Link from US grids to EDG grids
20ltltlt Grid Hype gtgtgt
21- Example The pejorization of Cluster
- Speaking loosly, Cray 1 was a cluster!
- DEC (? Windows NT)
- Beowulf clusters
- Sometimes cluster means nothing at all!
- Sometimes grid just means a
- distributed cluster not really a Grid.
- ? Keep in mind Fosters three criteria!
22Big Science
- (The future of most science)
23Enterprise Science more and more scientific
communities are finding themselves organizing
into what are effectively enterprises. Grid
computing as the foundation of enterprise activity
reveals a symmetry with industry and commerce a
symmetry that will benefit both.
24(No Transcript)
25(No Transcript)
26Natural Grid Applications
- High-energy elementary particle and Nuclear
Physics (HENP) - Distributed image processing
- Astronomy
- Biological/biomedical research e.g., pathology
- Earth and Planetary Sciences
- Military applications e.g., space surveillance
- Engineering simulations ? NEES Grid
- Distributed event simulations
- Military applications e.g., SF Express
- Medicine distributed, immersive patient
simulations ? Project Touch - Biology complete cell simulations
27(No Transcript)
28(No Transcript)
29(No Transcript)
30(No Transcript)
31Natural Grid Applications
- High-energy elementary particle and nuclear
physics (HE(N)P) - Distributed image processing
- Astronomy
- Biological/biomedical research e.g., pathology
- Earth and Planetary Sciences
- Military applications e.g., space surveillance
- Engineering simulations ? NEES Grid
- Distributed event simulations
- Military applications e.g., SF Express
- Medicine distributed, immersive patient
simulations ? Project Touch - Biology complete cell simulations
32(No Transcript)
33Natural Grid Applications
- High-energy elementary particle and nuclear
physics (HE(N)P) - Distributed image processing
- Astronomy
- Biological/biomedical research e.g., pathology
- Earth and Planetary Sciences
- Military applications e.g., space surveillance
- Engineering simulations ? NEES Grid
- Distributed event simulations
- Military applications e.g., SF Express
- Medicine distributed, immersive patient
simulations ? Project Touch - Biology complete cell simulations
34(No Transcript)
35Grid Applications (continued)
- Data mining and knowledge management
- Economics and Finance
- Bioinformatics
- Genomics and proteomics
- Pathology, virology
36And of course
37Applications not mentioned
- Some very important applications with significant
on-campus programs - computational fluid dynamics
- biological physics analyses e.g., phylogenetics
- weather forecasting
- are not mentioned because they are
super-computer-bound computations, (possibly)
not effectively distributable on a Grid. - HOWEVER Dissemination of their data and
knowledge products clearly demands Grid
technologies data bases, remote visulatization,
etc - ? No field will be un-touched by The Grid.
38At the University of New Mexico
- CERIA ? A perfect application on-campus Grid
first, then the world - MSB
- Sevilleta LTER
- NEON National Ecological Observatory Network
- much more!
- BIRN (Biomedical Informatics Research Network)
- UNM Center for Signal Transduction ? complete
cell simulations - Earth and Planetary Sciences
- HENP PHENIX and ATLAS experiment grids //
accelerator simulations - The Deep Lens Survey National Virtual
Observatory (NVO) LOFAR - The UNLV connection nuclear waste remediation
studies - The NCSA Alliance and Access Grid relationships
- and more!
39Grid Application Processing requirementsQuestion
s to ask
- How much raw computing (CPU) horsepower?
- What kind of computing (e.g., massively parallel
or not) - What are the data base parameters
- number and complexity
- geographic and administrative distribution
- data acquisition, population, and access rates
- ? Network QoS requirements
- total active size
- backup requirements
- Other QoS requirements
- Vendor-supplied or in-house service?
40Processing requirements
- Two examples
- Example 1 High-energy Nuclear Physics
- 10s of petabytes of data per year
- 10s of teraflops of distributed CPU power
- Comparable to todays largest supercomputers
41(No Transcript)
42(No Transcript)
43(No Transcript)
44(No Transcript)
45(No Transcript)
46(No Transcript)
47(No Transcript)
48(No Transcript)
49(No Transcript)
50CC-F
50 (300) TB/yr
CC-J
10 TB/yr
BNL
analysis, simulation, and reconstruction of
simulated data
analysis and reconstruction
5 TB/yr
10 TB/yr
WI
UNM simulation farm
off-site analysis
51And Future Projects will be EVEN BIGGER!
52Processing requirements
- But this is just the beginning!
- Example 2 data mining in the biosciences
53(No Transcript)
54Biological Databases Complex interdependencies
- Domino-effect in data publishing
- Efficiently keep many versions
GERD
TRRD
EpoDB
BEAD
BEAD
Swissprot
Swissprot
GAIA
EMBL
GenBank
Transfac
Transfac
DDBJ
(Yong Zhao, University of Chicago)
55Data Mining Example
56The Future
57and the role of Internet 2.
- It is clear that advanced networking will play a
critical role in the development of an intergrid
and its eventual evolution into The Grid - Broadband capacity
- Advanced networking protocols
- Well-defined, finely graded, clearly-costed high
Qualities of Service
58(No Transcript)
59(No Transcript)
60(No Transcript)
61Connectivity of the web one can pass from any
node of IN through SCC to any node of OUT.
Hanging off IN and OUT are TENDRILS containing
nodes that are reachable from portions of IN, or
that can reach portions of OUT, without passage
through SCC. It is possible for a TENDRIL
hanging off from IN to be hooked into a TENDRIL
leading into OUT, forming a TUBE -- a passage
from a portion of IN to a portion of OUT without
touching SCC.
62(No Transcript)
63In other wordsbarely predictable
- But no doubt inevitable, disruptive,
transformative, - and very exciting!