V. Milutinović, G. Rakocevic, S. Stojanović, and Z. Sustran University of Belgrade Oskar Mencer...
-
Upload
martin-kelly -
Category
Documents
-
view
215 -
download
0
Transcript of V. Milutinović, G. Rakocevic, S. Stojanović, and Z. Sustran University of Belgrade Oskar Mencer...
V. Milutinović, G. Rakocevic, S. Stojanović, and Z. Sustran
University of Belgrade
Oskar MencerImperial College, London
Oliver Pell Maxeler Technologies, London and Palo Alto
Michael FlynnStanford University, Palo Alto
1/60
For Big Data algorithms and for the same hardware price as before, achieving:
a) speed-up, 20-200 b) monthly electricity bills, reduced
20 timesc) size, 20 times smaller
2/60
1. BigData2. WORM3. Tolerance to latency4. Over 95% of run time in loops5. Reusability of the data6. Skills
3/60
Absolutely all results achieved with:
a) all hardware produced in Europe, specifically UK
b) all software generated by programmers
of EU and WB
4/60
ControlFlow (MultiFlow and ManyFlow): Top500 ranks using Linpack (Japanese
K,…)
DataFlow: Coarse Grain (HEP) vs. Fine Grain
(Maxeler)
5/60
Compiling below the machine code level brings speedups;also a smaller power, size, and cost.
The price to pay:The machine is more difficult to program.
Consequently:Ideal for WORM applications :)
Examples using Maxeler:GeoPhysics (20-40), Banking (200-1000, with JP Morgan
20%), M&C (New York City), Datamining (Google), …
6/60
7
8/60
9/60
10
11
Assumptions: 1. Software includes enough parallelism to keep all cores busy 2. The only limiting factor is the number of cores.
tGPU = N * NOPS * CGPU*TclkGPU / NcoresGPU
tCPU = N * NOPS * CCPU*TclkCPU /NcoresCPU
tDF = NOPS * CDF * TclkDF + (N – 1) * TclkDF / NDF
12/60
DualCore?
Which way are the horses going?
13/60
Is it possibleto use 2000 chicken instead of two horses?
?==
14/60
What is better, real and anecdotic?
2 x 1000 chickens (CUDA and rCUDA) 15/60
How about 2 000 000 ants?
16/60
Dat
a
Marmalade
Big Data Input Results
17/60
Factor: 20 to 200
MultiCore/ManyCore
Dataflow
Machine Level Code
Gate Transfer Level
18/60
Factor: 20
MultiCore/ManyCore
Dataflow
19/60
Factor: 20
Data Processing
Process ControlData Processing
Process Control
MultiCore/ManyCore
DataFlow
20/60
MultiCore:Explain what to do, to the driverCaches, instruction buffers, and predictors needed
ManyCore:Explain what to do, to many sub-driversReduced caches and instruction buffers needed
DataFlow:Make a field of processing gates: 1C+2nJava+3JavaNo caches, etc. (300 students/year: BGD, BCN, LjU,
ICL,…)
21/60
MultiCore:Business as usual
ManyCore:More difficult
DataFlow:Much more difficultDebugging both, application and configuration
code
22/60
MultiCore/ManyCore:Several minutes
DataFlow:Several hours for the real hardwareFortunately, only several minutes for the simulatorThe simulator supports
both the large JPMorgan machineas well as the smallest “University Support” machine
Good news:Tabula@2GHz
23/60
24/60
MultiCore:Horse stable
ManyCore:Chicken house
DataFlow:Ant hole
25/60
MultiCore:Haystack
ManyCore:Cornbits
DataFlow:Crumbs
26/60
27/60
Small Data: Toy Benchmarks (e.g., Linpack)
28/60
Medium Data (benchmarks favorising NVidia,compared to Intel,…)
29/60
Big Data
Revisiting the Top 500 SuperComputer BenchmarksOur paper in Communications of the ACM
Revisiting all major Big Data DM algorithmsMassive static parallelism at low clock frequencies
Concurrency and communicationConcurrency between millions of tiny cores difficult,
“jitter” between cores will harm performance at synchronization points
Reliability and fault tolerance10-100x fewer nodes, failures much less often
Memory bandwidth and FLOP/byte ratioOptimize data choreography, data movement,
and the algorithmic computation
30/60
Maxeler Hardware
CPUs plus DFEsIntel Xeon CPU cores and up to
4 DFEs with 192GB of RAM
DFEs shared over Infiniband Up to 8 DFEs with 384GB of RAM and dynamic allocation
of DFEs to CPU servers
Low latency connectivityIntel Xeon CPUs and 1-2 DFEs with up to six 10Gbit Ethernet
connections
MaxWorkstationDesktop development system
MaxCloudOn-demand scalable accelerated compute resource, hosted in London
3131/60/60
1. Coarse grained, stateful: Business– CPU requires DFE for minutes or hours
2. Fine grained, transactional with shared database: DM– CPU utilizes DFE for ms to s– Many short computations, accessing common database data
3. Fine grained, stateless transactional: Science (FF)– CPU requires DFE for ms to s– Many short computations
3232/60/60
Major Classes of Algorithms, from the Computational Perspective
• Long runtime, but:• Memory requirements
change dramatically based on modelled frequency
• Number of DFEs allocated to a CPU process can be easily varied to increase available memory
• Streaming compression• Boundary data exchanged
over chassis MaxRing
3333/60/60
Coarse Grained: Modeling
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
1 4 8
Equi
vale
nt C
PU c
ores
Number of MAX2 cards
15Hz peak frequency
30Hz peak frequency
45Hz peak frequency
70Hz peak frequency
0
10
20
30
40
50
60
70
80
0 10 20 30 40 50 60 70 80Peak Frequency (Hz)
Timesteps (thousand)
Domain points (billion)
Total computed points (trillion)
• DFE DRAM contains the database to be searched• CPUs issue transactions find(x, db)• Complex search function
– Text search against documents– Shortest distance to coordinate (multi-dimensional)– Smith Waterman sequence alignment for genomes
• Any CPU runs on any DFE that has been loaded with the database– MaxelerOS may add or remove DFEs
from the processing group to balance system demands– New DFEs must be loaded with the search DB before use
3434/60/60
Fine Grained, Shared Data: Monitoring
• Analyse > 1,000,000 scenarios• Many CPU processes run on many DFEs
– Each transaction executes on any DFE in the assigned group atomically
• ~50x MPC-X vs. multi-core x86 node
3535/60/60
Fine Grained, Stateless: The BSOP Control
CPU DFE Loop over instrumentsLoop over instruments
Random number generator and
sampling of underliers
Random number generator and
sampling of underliers
Price instruments using Black
Scholes
Price instruments using Black
Scholes
Tail analysis on CPU
Tail analysis on CPU
CPU DFE Loop over instrumentsLoop over instruments
Random number generator and
sampling of underliers
Random number generator and
sampling of underliers
Price instruments using Black
Scholes
Price instruments using Black
Scholes
Tail analysis on CPU
Tail analysis on CPU
CPU DFE Loop over instrumentsLoop over instruments
Random number generator and
sampling of underliers
Random number generator and
sampling of underliers
Price instruments using Black
Scholes
Price instruments using Black
Scholes
Tail analysis on CPU
Tail analysis on CPU
CPU DFE Loop over instrumentsLoop over instruments
Random number generator and
sampling of underliers
Random number generator and
sampling of underliers
Price instruments using Black
Scholes
Price instruments using Black
Scholes
Tail analysis on CPU
Tail analysis on CPU
DFE Loop over instrumentsLoop over instrumentsCPUMarket and instruments data
Random number generator and
sampling of underliers
Random number generator and
sampling of underliers
Price instruments using Black
Scholes
Price instruments using Black
ScholesInstrument values
Tail analysis on CPU
Tail analysis on CPU
3636/60/60
Selected Examples
37373737/60/60
Performance of one MAX2 card vs. 1 CPU core
Land case (8 params), speedup of 230x
Marine case (6 params), speedup of 190x
The CRS Results
CPU Coherency MAX2 Coherency
3838/60/60
Seismic Imaging
• Running on MaxNode servers- 8 parallel compute pipelines per chip- 150MHz => low power consumption!- 30x faster than microprocessors
An Implementation of the Acoustic Wave Equation on FPGAs T. Nemeth†, J. Stefani†, W. Liu†, R. Dimond‡, O. Pell‡, R.Ergas§
†Chevron, ‡Maxeler, §Formerly Chevron, SEG 20083939/60/60
4040
• DM for Monitoring and Control in Seismic processing • Velocity independent / data driven method
to obtain a stack of traces, based on 8 parameters– Search for every sample of each output trace
Trace Stacking: Speed-up 217P. Marchetti et al, 2010
parameters( emergence angle & azimuth
Normal Wave front parametersKN,11; KN,12 ; KN22
NIP Wave front parameters( KNip,11; KNip,12 ; KNip22 )
hHKHhmHKHmmw TzyNIPzy
TTzyNzy
TT
0
0
2
00
2 22
v
t
vtthyp
4242/60/60
4343
4444
4545
This is about algorithmic changes, to maximize
the algorithm to architecture match:Data choreography, process modifications,
anddecision precision.
The winning paradigm of Big Data ExaScale?
4646/60/60
Conclusion: Nota Bene
47/8
The TriPeak
Siena+ BSC+ Imperial College + Maxeler+ Belgrade
46/60
48/8
The TriPeakMontBlanc = A ManyCore (NVidia) + a MultiCore (ARM)Maxeler = A FineGrain DataFlow (FPGA)
How about a happy marriage?MontBlanc (ompSS) and Maxeler (an accelerator)
In each happy marriage,it is known who does what :)
The Big Data DM algorithms:What part goes to MontBlanc and what to Maxeler?
48/60
49/8
Core of the Symbiotic SuccessAn intelligent DM algorithmic scheduler,partially implemented for compile time,and partially for run time.
At compile time:Checking what part of code fits where(MontBlanc or Maxeler): LoC 1M vs 2K vs 20K
At run time:Rechecking the compile time decision,based on the current data values.
49/60
50/850/850
51/851/8
Maxeler: Teaching (Google: prof vm)TEACHING, VLSI, PowerPoints, Maxeler:
Maxeler Veljko Explanations, August 2012Maxeler Veljko Anegdotic, Maxeler Oskar Talk, August 2012Maxeler Forbes ArticleFlyer by JP MorganFlyer by Maxeler HPCTutorial Slides by Sasha and Veljko: Practice (Current Update)Paper, unconditionally accepted for Advances in Computers by ElsevierPaper, unconditionally accepted for Communications of the ACMTutorial Slides by Oskar: Theory (7 parts)Slides by Jacob, New YorkSlides by Jacob, AlabamaSlides by Sasha: Practice (Current Update)Maxeler in MeteorologyMaxeler in MathematicsExamples generated in Belgrade and Worldwide
THE COURSE ALSO INCLUDES DARPA METHODOLOGY FOR MICROPROCESSOR DESIGN, with an example
51/60
52/852/8© H. Maurer52
53/853/8
Maxeler: Research (Google: good method)
Structure of a Typical Research Paper: Scenario #1[Comparison of Platforms for One Algorithm]Curve A: MultiCore of approximately the same PurchasePriceCurve B: ManyCore of approximately the same PurchasePriceCurve C: Maxeler after a direct algorithm migrationCurve D: Maxeler after algorithmic improvementsCurve E: Maxeler after data choreographyCurve F: Maxeler after precision modifications
Structure of a Typical Research Paper: Scenario #2[Ranking of Algorithms for One Application]CurveSet A: Comparison of Algorithms on a MultiCoreCurveSet B: Comparison of Algorithms on a ManyCoreCurveSet C: Comparison on Maxeler, after a direct algorithm migrationCurveSet D: Comparison on Maxeler, after algorithmic improvementsCurveSet E: Comparison on Maxeler, after data choreographyCurveSet F: Comparison on Maxeler, after precision modifications
53/60
54/854/8© H. Maurer54
55/855/8
Maxeler Tutorials
Google:
IEEE HiPeak, Berlin, Germany, January 2013
ACM iSAC, Coimbra, Portugal, March 2013
IEEE MECO, Budva, Montenegro, June 2013
ACM ISCA, Tel Aviv, Israel, June 2013
55/60
56/856/8
Maxeler Research in Serbia
KG: Blood Flow, Prof. Filipovic and Tijana Djukic
NS: Combinatorial Math, Prof. Senk and Ivan Stanojevic
MISANU: The SAT Math, Ognjanovic
ETF: Meteorology, Radomir Radojicic and Marko Stankovic
ETF: Physics (Gross Pitaevskii 3D real), Lena Parezanovic
ETF: Physics (Gross Pitaevskii 3D imaginary), Sasa Stojanovic
56/60
57/857/8
Maxeler: DAFNE
MISANU, IMP, KG, NS, BSC, UPV, U of Siena, U of Roma,IJS, FRI,IRB, QPLAN, Bogazici, U of Istanbul,U of Bucharest, U of Arad,U of Tuzla, Technion, Maxeler Israel, IPSI
57/60
58/858/8
Maxeler: Advances in Computers
MISANU, IMP, KG, NS, BSC, UPV, U of Siena, U of Roma,IJS, FRI,IRB, QPLAN, Bogazici, U of Istanbul,U of Bucharest, U of Arad,U of Tuzla, Technion, Maxeler Israel, IPSI
58/60
59/859/8© H. Maurer59 [email protected]
Q&A