8/9/2019 Defining Grand Challenges
1/28
April 13, 2010 CDAC Symposium High
Performance Computing
11April 13, 2010
D ef in in g Gr a n d Ch a lle n g e s
Pr a do sh K R o y FIETE
N a t io n a l S ym p o s iu m o n H ig h P e r fo r m a n ce Co m p u t in g
S p o n s o r e d b y
Ce n t r e f o r D ev e lo p m e n t o f A d v a n c e d Co m p u t i n g , Pu n e
B e n g a l E n g i n e er i n g & S c ie n c e U n i v e r s i t y , S h i b p u r
4 t h March ,2010
8/9/2019 Defining Grand Challenges
2/28
April 13, 2010 CDAC Symposium High
Performance Computing
22April 13, 2010
S t a t u t o r y D ecla r a t io n s
P e n t i u m , X eon a r e t h e r e g is t e r e d t r a d e m a r k s o f In t e l C o r p o r a t i o n .
I B M P o w e r i s t h e r e g i st e r e d t r a d e m a r k o f I B M Co r p .
BSD i s t h e r e g i s t er e d t r a d e m a r k o f U U n e t Te ch n o l o g ie s I n c .
UNIX i s t h e r e g is t e r e d t r a d e m a r k o f Op e n G r o u p .
Mac i s t h e r e g is t e r e d t r a d e m a r k o f A p p l e Co r p .
L inux
i s reg is te red a t Free So f tw are Found a t ion [FSF] W i n d o w s i s t h e r e g i s t er e d t r a d e m a r k o f M i c r o s of t C o r p o r a t i on .
M y r i n e t a n d M y r i c om a r e r e g is t e r e d t r a d e m a r k s o f M y r i co m I n c .
I n f i n i b a n d i s d e v e l o p e d b y C o m p a q , IB M , H P , In te l , Microsof t ,Sun
Co n n e t X a n d I n f i n i H o s t a r e t h e r e g is t e r e d t r a d e m a r k s o f M e ll a n o x Tec h n o l o g ie s L t d .
P o r t a b le B a t c h S y s t e m [ PB S ] i s d e v e lo p e d a t N A S , N A S A .
NAS Para l le l Benchm ark Su i te [NPB] i s dev e loped a t NAS A Am es Research Cen tr e , Ca l if .
M P I i s d e v e lo p e d a t A r g o n n e N a t i o n a l La b o r a t o r y , U n i v e r s it y o f Ch i c a g o .
P V M i s d e v e lo p e d b y O a k R i d g e N a t i on a l La b o r a t o r y
O p e n M P i s t h e r e g is t e r e d t r a d e m a r k o f Op e n M P O r g a n i z a t i o n .
CO N D OR i s d e v e l o p e d a t U n i v e r s i t y o f W i s co n s i n - M a d i s o n , W i s co n s i n
G lo b u s To o l K i t w a s d e v e l o p e d b y Te a m o f Co m p u t e r S c i e n t i st s f r o m A N L/ E P CC/ N CS A / U S C is n o w
ava i lab le un der Genera l Pub l ic L icense
8/9/2019 Defining Grand Challenges
3/28
April 13, 2010 CDAC Symposium High
Performance Computing
33April 13, 2010
S ch e m e Of Pr e s en t a t io n
D ef in i n g G r a n d Ch a lle n g e s & N a t u r e o f t h e Ph y s ic a l W o r l d
A l gor i t hm s & Com pl e x it y [P # NP ]
H i g h P er f or m a n ce Co m p u t i n g : H a r d w a r e / S o f t w a r e I n t e r fa c e M a r k B a k e r & I E EE W h i t e Pa p e r o n C lu s t e r Co m p u t in g [ 20 0 1]
Pa r a l le l Pr o g r a m m in g E n v i r o n m e n t s [ M P I ,PV M , Op e n M P]
H igh P r e c is i on Com pu t a t i on
N a t i o n a l P o li cy f o r S u p e r co m p u t in g : N S F 0 7 2 8 R e p o r t
Fe y n m a n & T h eo r y Q u a n t u m Co m p u t i n g .
Conc lus ion
8/9/2019 Defining Grand Challenges
4/28
8/9/2019 Defining Grand Challenges
5/28
April 13, 2010 CDAC Symposium High
Performance Computing
55April 13, 2010
P h i lo s o p h y o f H P C : G r a n d Ch a lle n g e s f r o m
70 0 B C t o M o d e rn D a y s
Grand Challenges in Anci ent Civilizations
P la n e G eo m e t r y
A s t r o n o m y
S h a p e a n d R a d i u s o f E a r t h [ E r a s t o s t h e n e s 2 4 0 B C]
D is t a n c e b e t w e e n S u n a n d E a r t h
S o la r a n d Lu n a r E cl ip s e s
L u n a r C a l en d a r
Naviga t ionT r i go n o m et r i c Tab l e s
S u r y a S i d d h a n t a [ 70 0 B C ? ]
Alm ages t [110 AD] , S idd ha n t a S i r om an i [1120 AD]
N u m b e r S y st e m
B ab y lo n i an [Sexag es i m a l ] [20 0 0 B C ]
Hi n d u -Ara b i c [Dec im a l ] [ ? 70 0 B C]
I o n i a n
A r i t h m e t i c
3 D Ge o m e t r y o f P y r a m i d s
Algebra
Met a ls & Alloys
Med ic in a l Ut i l ity o f P lan tsBabylonian Plimpton 322
8/9/2019 Defining Grand Challenges
6/28
April 13, 2010 CDAC Symposium High
Performance Computing
6
H in d u T r ig o n o m e t r y 70 0 B C(?)
S in e o f 3 = 2 2 5 / 3 4 3 8 [ 0 . 0 6 5 44 5 0 2 6 ]
[Ac tua l Va lue i s 0 .06 540 3129 ]
8/9/2019 Defining Grand Challenges
7/28
8/9/2019 Defining Grand Challenges
8/28
April 13, 2010 CDAC Symposium High
Performance Computing
88
April 13, 2010
N at ure of Physical W orld2500BC Babylonian/Hindu/Egyptian/Mayan Civilizations] [Astronomy]
500BC : Pythagorus [ Geometry]450BC : Democri tus / Plato
300BC-200BC : Aris tarchus / Erastosthenes/Euclid [Geometry/Trigonometry/]100BC- AD : Antikythera Machine
0AD-100 AD : Heron o f Alexandria [Gear Trains]100AD 200AD : Ptolemy [Astronomy/Geography]
400AD-500 AD : Destruction of Alexandrian Library [Dark Age]Da Vinci/[Italian Renaissance][Art/Science/]
500AD-1500 AD : Poignant Lost Opportuni ty for Human Species [Carl Sagan]
1500AD- 1600 AD : Copernicus/ Tyco Brahe/Keppler/Galileo [ Solar syst em l/Dynamics]1600AD-1700AD : Isaac Newton/Huygens/Leibnitz [Physical Science]1700 AD- 1900 AD : Michael Faraday/Maxwell/ Hertz/ J C Bose[Physical Science]
1900AD 1930 AD : Einstein/Max Planck/Bohr/S N Bose/Raman/Saha/ Heisenberg/Schroedinger /Pauli/Fermi
Lise Meitner/Paul Dirac/ Lev Landau/Feynman[ Physical Sciences]1945 AD 1995 AD : Astrophysics/Quantum Electrodynamics/Particle Physics //BioSciences
1995 AD- Present Day : Data Analysis/Visualization/Simulation of Macro Phenomena/Bioinformatics/Nonlinear Phenomena &c
2025AD (?) - ? : Quantum Computers & NP Problems
8/9/2019 Defining Grand Challenges
9/28
April 13, 2010 CDAC Symposium High
Performance Computing
99April 13, 2010
Ph ilo sop h y o f HP C : 20 t h /21s t Ce n t u r y Gr a n d Ch a l le n ge s
Yea r Sys t em Sp e ed Sys te m Me m o r y Gr a n d Ch a llen ge
19 8 0 10 0 M F LOP S 10 M B 4 8 H o u r W e a t h e r M o d e l in g
19 9 0 1 GFLOP S 5 GB -10 GB + 3 D P l a s m a M o d e lin g / 7 2 H o u r W e a t h e r
19 9 3 10 0 GFLP OS > 10 GB + P h a r m a c e u t i c a l D e s ig n / S t r u c t u r a l Bi o lo g y
G e o p h y s ic a l I m a g in g / R e s e r v o ir S im u l a t i o n
19 9 5 1TFLOP S 10 3GB + Globa l C lima te Cha nge/ CFD/Ocea n
Ci r c u l a t io n / S e m i c o n d u c t o r M o d e l i n g /
Co m p u t e r V is i o n / B io i n f o r m a t i cs /
Vehic le Dynam ics
20 0 4 * 10 6 GB + NASA Scie n ce M is s io n Dir ec to r a t e
Solar Sys tem Science
Or igin of LifeB ig B a n g a n d D a r k E n e r g y ..
> 2 0 10 1 P FLOP S 10 9 GB
NASA Institute of Advanced Conceptshttp://www.niac.usra.edu
8/9/2019 Defining Grand Challenges
10/28
April 13, 2010 CDAC Symposium High
Performance Computing
1010April 13, 2010
Ph ilo sop h y o f HP C : 20 t h /21s t Ce n t u r y G r a n d Ch a lle n ge s
Yea r Sys t em
R e q u i r e m e n t s
Ag en cy Gran d Ch a l len g e
19 9 5 1 TFLOP NSF A e r o s p a c e
Molecu lar B io logy
Space Sc ience
E n v i r o n m e n t a l M o d e l i n g
DOE E n e r g y
E n v ir o n m e n t a l M o n i t o r i n g & P r e d i ct i on
Co m p u t a t io n a l C h e m i s t r y
NASA G a l a xy F o r m a t i o n
Convec t ive Tur bu lence
H e l io s p h e r i c D yn a m i cs
NIH Molecu lar B io logy
B io m e d i c a l I m a g in g
NIST P r o d u c t D e s i gn & P r o c e s s O p t i m i z a t io n
NOAA W e a t h e r F o r e c a st in g
Grand Challenges N ITR D Govt of USThird Printing November,2006
8/9/2019 Defining Grand Challenges
11/28
April 13, 2010 CDAC Symposium High
Performance Computing
11
11
April 13, 2010
Algor ith m s & Com p lex ityThe Class ica l Theor y
Comparis on n**2 & nlgn [FFT]
-20000
020000
40000
60000
80000
100000
120000
140000
160000
180000
n
n2,nlgn
O( 2 n ) ~ 366 Cen tu r i es @1979 Process o r
O(3 n ) ~ 1.31 x 10 13 Centu r i es@ 1979 Processo r
F o r n = 6 0
Reference 1.D a s g u p t a , P a p a d i m i t r i o u , V a z ir a n i : A lg o r i t h m s
2 . A b h i r a m R a n a d e :/ / w w w . y o u t u b e .c om
[D e s ig n a n d A n a l y s i s o f A lg o r i t h m s]
8/9/2019 Defining Grand Challenges
12/28
April 13, 2010 CDAC Symposium High
Performance Computing
12
I n t e l M icr o p r o ce s so r P e r fo r m a n ce
April 13, 2010 12
Com puter Architecture 7thEdition : W illiam Stallings
8/9/2019 Defining Grand Challenges
13/28
April 13, 2010 CDAC Symposium High
Performance Computing
13
S u p e r co m p u t er A r ch it ect u r e
19 6 5 - 2 0 0 0
8/9/2019 Defining Grand Challenges
14/28
April 13, 2010 CDAC Symposium High
Performance Computing
1414April 13, 2010
H P C : H a r d w a r e / So ft w a r e I n t e r fa ce
IEEE TFCC Clus te r Com pu t ing W hi te Paper [200 0]
8/9/2019 Defining Grand Challenges
15/28
April 13, 2010 CDAC Symposium High
Performance Computing
1515April 13, 2010
H P C : H a r d w a r e / So ft w a r e I n t e r fa ce
IEEE TFCC Cluster Computing White Paper [2000]
8/9/2019 Defining Grand Challenges
16/28
April 13, 2010 CDAC Symposium High
Performance Computing
1616April 13, 2010
Obvious huge interest in Clusters, seemed natural toset up a focussed group in this area.
A Cluster Computing Task Force was proposed to the
IEEE CS, approved and started operating in February1999 been going just over 2 years.
Task Forces: Expected to have a finite life (
8/9/2019 Defining Grand Challenges
17/28
April 13, 2010 CDAC Symposium High
Performance Computing
1717April 13, 2010
H P C Tr e n d s : P r o ce s s o r F a m ily
http://www.top500.org
In te l EM64T
Int el IA 32
8/9/2019 Defining Grand Challenges
18/28
April 13, 2010 CDAC Symposium High
Performance Computing
1818April 13, 2010
H P C Tr e n d s : O p e r a t in g Sys t e m s
L i n u x
UNIX
8/9/2019 Defining Grand Challenges
19/28
April 13, 2010 CDAC Symposium High
Performance Computing
1919April 13, 2010
Pa r a l le l E n v ir o n m e n t [GPL ]
H igh l eve l Lan gua ges : FOR TRAN 77 , C , C++
[e .g . f77,gcc com p i le r s in Lin u x]
P a r a l le l P r o gr a m m in g E n v ir o n m e n t s
M P I [ M e s s a ge P a s s in g I n t e r f a ce ] [ Ar o go n n e N a t io n a l La b ]
[ de fa c t o st a n da r d fo r D M M / D S M Ar c h i te c tu r e s ]PVM [Pa r a l le l Vi r tu a l Machin e ] [Oak Ridge Na t ion a l Lab ]
O p e n M P [ F o r S M P Ar c h i t ect u r e s ]
N um e r i ca l Li b r a r ie s
Nu m er i ca l Rec ipe s : The Ar t o f Sc ien t if ic Com pu t ing
W H Pr es s , Sau l Teu k o l s k y , Ve l le r i n g , F lan n e r y [CUP , 19 8 6 ]
I M S L [ Vis ua l N um e r i cs I nc . ] ,
N AG [ N um e r i ca l Algo r i t hm G r ou p]
Clus t e r M a n a ge m e n t S ys t e m
8/9/2019 Defining Grand Challenges
20/28
April 13, 2010 CDAC Symposium High
Performance Computing
2020April 13, 2010
N A S A G o d d a r d S p a ce Flig h t Ce n t r eT h e C lu s t e r S u p e r co m p u t e r
IBM Bl u e Gen e
1999 The Stone Souper ORNL
8/9/2019 Defining Grand Challenges
21/28
April 13, 2010 CDAC Symposium High
Performance Computing
2121
Sim u la t ion o f Black H o le Mer ge rAn Exa m pl e o f S im u l a t i on
M e r g e r o f B l a c k H o l es a n d R i p p le s i n S p a c e T i m e
Co u r t e s y : N a t i o n a l Ce n t r e f o r S u p e r c o m p u t i n g A p p l ic a t io n s , U n i v e r s i t y o f I ll in o i s .
8/9/2019 Defining Grand Challenges
22/28
April 13, 2010 CDAC Symposium High
Performance Computing
2222April 13, 2010
A s t h e S u n r i s es o v e r E a s t e r n E u r o p e , t h e i n s t a n t a n e o u s n e t e co s y s t e m e x ch a n g e ( N E E) o f CO 2 is s h o w n i n
t h e E a s t e r n H e m i s p h e r e . S t r o n g u p t a k e i s s h o w n i n g r e e n - t o -w h i t e co l o r s a n d i s s t r o n g e s t in t h e t r o p i cs .
A n e t r e le a s e o f CO 2 t o t h e a t m o s p h e r e i s s h o w n i n r e d -t o -w h i t e c o lo r s a n d i s s t r o n g e s t o v e r t h e Co n g o ,
w h e r e t h e S u n i s n o t s h i n i n g . Th i s im a g e w a s p r o d u c e d f r o m a C -LA M P s i m u l a t io n p e r f o r m e d a s p a r t o f a
Sc iDAC-2 pro jec t u s ing NCCS sup ercom pu ters . DOE, SciDac, ORNL, USA
S im u la t io n o f Ne t E co sys t em E xch an ge a t
SciDAC ,ORNL ,20 0 9
8/9/2019 Defining Grand Challenges
23/28
April 13, 2010 CDAC Symposium High
Performance Computing
2323
H ig h Pr e cis io n Co m p u t in g
April 13, 2010
S EEEEEEEEEEE FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF 0 1 11 12 63The value V represented by the word may be determined as follows:
If E=2047 and F is nonzero, then V=NaN ("Not a number")If E=2047 and F is zero and S is 1, then V=-InfinityIf E=2047 and F is zero and S is 0, then V=Infinity
If 0
8/9/2019 Defining Grand Challenges
24/28
April 13, 2010 CDAC Symposium High
Performance Computing
2424
Ne cess i ty for
N a t io n a l Su p e r co m p u t in g P o licy
April 13, 2010
NSF VISION FOR 20 10 [ vide NSF0 728 ]
H PC Sys t e m a n d Se r v ic e s fo rP e t a -sc a l e Sc i en t i f ic Com pu t i ng
2 . At Le a s t o n e S y s t e m ca p a b le o f s u s t a in i n g 10 15 FLOPS
3 . D e ve lo p m e n t a n d M a in t e n a n ce o f P e r f o r m a n ce To o ls ,
S ys t e m S o ft w a r e a n d F u n d a m e n t a l Algo r i th m s.
4 . D eve lo p m e n t a n d M a in t e n a n ce o f Po r t a b le Sca l a b l e
App lic a t ion So ft wa r e a n d St a n d a r d s [ e .g CDAC ONAMA].
5. De ve lo p m e n t o f N a t i o n a l D ig it a l D a t a F r a m e w o r k .
6 . D eve lo p m e n t o f N e tw o r k e d R e so u r c e s a n dVir t u a l Orga n iz a t ion s.
7. Us in g Cyb e r -in fr a s t r u c t u r e fo r E n h a n c in g L e a r n in g.
8/9/2019 Defining Grand Challenges
25/28
April 13, 2010 CDAC Symposium High
Performance Computing
2525
H PC : A B u s in ess Persp ect i v e
H igh P e r fo r m a n c e Ap p lic a t io n S o ft w a r e- S e is m ic Im a g in g , R e s e r v o ir S im u l a t io n
- W e a t h e r M o d e lin g & F o r e c a s tin g
- S e m ico n d u c t o r M o d e lin g
- Ae r o s p a c e E n g in e e r in g & S im u l a t io n- D a ta M in i n g
- Visu a l iz a t ion &c .
H igh P e r fo r m a n ce H a r d w a r e Ar c h i t ec tu r e H igh P e r f o r m a n c e S ys t e m S o ft w a r e D e ve lo p m e n t / S e r vi ce s H igh Avai la b i li ty Netw or k : Design & Ser vices
Da t a Ce n t r e Ar c h i t e c tu r e : De s i gn & Se rv ic e s ad infinitum.
8/9/2019 Defining Grand Challenges
26/28
April 13, 2010 CDAC Symposium High
Performance Computing
2626
Fu t u r e o f S u p e r co m p u t in g
Fey n m a n & Qu a n t u m Co m p u t in g
April 13, 2010
Pr im e Fa c t o r iz a t i o n [P et e r W S h o r , 19 8 7 ]
Cla ss ica l A lgor i thm O (n ) = e xp [2 (lg n ) 1/ 3 ( lg lg n )2/ 3]
S h o r s A lg o r i th m O (n ) [198 7] = ( lg n ) 3[ a r x i v : q u a n t - p h / 9 5 0 8 0 2 7 v 2 / 25 th Ja n 199 6]
S e a r c h in g U n s o r t e d D a t a b a s e [L K Gr o v e r 19 9 6 ] ,
Cla ss ica l A lgor i thm O (n ) n / 2 O p er a t i o n s
Grov e r s A l gor i t h m O ( n ) = n O p e r a t i o n[a r x i v . o r g :q u a n t -p h / 0 10 9 116 2 2 n d S e p t e m b e r , 20 0 1]
8/9/2019 Defining Grand Challenges
27/28
April 13, 2010 CDAC Symposium High
Performance Computing
2727April 13, 2010
Conc lus ion
T h e K n o w n is f in i t e , th e u n k n o w n in f in i t e ,in t e lle ct u a lly w e s t a n d o n a n is le t in t h e m id s t o f a ni ll im i t ab l e oc ean o f i ne x p li cab i li t y . O ur bu s i ne s s i ne v e r y g e n e r a t i o n is t o r e cla i m a l it t le m o r e la n d
- T H H u x l ey
8/9/2019 Defining Grand Challenges
28/28
April 13, 2010 CDAC Symposium High
Performance Computing
28
S e le ct e d R e fe r e n ce s
1. The Sleepwalkers, Arthur Koestler , Penguin Books,1969
2. Act of Creation, Arthur Koestler, Penguin Books, 1964
3. Cosmos , Car l Sagan, Bal lantine Books,1974
4. Surya Siddhnta , Swami Vijnanananda, Sanskrit Book Depot,Calcutta,2005
5. Atomic Physics & Human Knowledge, Niels Bohr, Braunschweig, 1985
6. The Science of Leonardo , Fri tjof Capra, Anchor Books,20077. Feynman Lectures on Physics , Vol 3, Addission Wesley Inc.,1964
8. Feynman Lectures on Computation , Perseus Books ,1996
9. Using MPI , W Gropp,E.Lusk , A.Skjellum, The MIT Press , 1994
10. Computer Architecture : Hardware Software Interface, Hennessey/Patterson ,
11. Parallel Computer Architecture : David Cueller, Jaswinder Singh, Anoop Gupta
12. Great Ideas in Theoretical Computer Science, 6.089, MIT Open Courseware,2008
.
April 13, 2010 28
T H A N K S