Pr

1
Multi-Programming and Scheduling Design for
Applications of Interactive SimulationJean-Loui
s Roch al.
http//moais.imag.fr
Louvre, Musée de lHomme Sculpture (Tête) Artist
Anonyme Origin Rapa Nui Easter Island Date
between the XIst and the XVth century Dimensions
1,70 m high
EVALUATION SEMINAR -RESEARCH THEME Num B"Grids
and high-performance computing"March 27-28, 2008
2
Staff and Skills

• 1/1/2005 Creation of MOAIS team1/1/2006
  Creation of INRIA team-project MOAIS

• Vincent Danjean MdC 9/2005
• Pierre-François Dutot MdC 9/2006
• Thierry Gautier CR
• Guillaume Huard MdC
• Grégory Mounié MdC
• Bruno Raffin CR INRIA
• Jean-Louis Roch MdC, Team leader
• Denis Trystram Prof
• Frédéric Wagner MdC 9/2006
• 1 Invited Prof. Alfredo Goldman USP Sao
  Paulo
• 19 PhD students, 1 engineer
• 14 PhDs defended since 2005

• Parallel algorithms programming
• Scheduling
• Interactive applications

3
Evolution of parallel programming

• Parallelism everywhere
• Distributed, Heterogeneous

MPSoC
Grids
Cluster
SMP
multi-core
GPU
MPI OpenMP
Cuda NVidia
MapReduce Google
TBB Intel
SPIRIT
Cilk CilkArts
Fortress Sun
4
MOAIS objective

• End-to-end parallel programming solutionsfor
  high-performance interactive computing with
  provable performances.
• optimization computational
  steering, VR embedded
• Performance is multi-objective

QAP/Nugent on Grid5000 PRISM, GSCOP, DOLPHIN
Streaming on MPSoCs ST
INRIA Grimage platform MOAIS, PERCEPTION,
EVASION
5
Approach

• To mutually adapt application and scheduling.
• Proactive/static to the platform the devices
  evolve gradually
• Online/dynamic to the execution context data
  and resources
• Tolerant to data variations, failures, other
  appli. perturbation,
• From algorithms to applications
• Scheduling and parallel programming schemes
• Programming interfaces and tools
• Target applications batch scheduling,
  combinatorial optimization, computational
  steering, stream encoding

6
Overview
Interactive application
Adaptive control of execution
M O A I S
model abstract representation
algorithm scheduling,
fault tolerance
Architecture
7
Research directions and achievements for
2005-2007

1. Scheduling
2. Interfaces for coordination
3. Adaptive algorithms
4. Interactive applications

8
1. Scheduling

• Objective A modeling of scheduling problems for
  adaptive applications
• Adaptable parallelism degree for efficient coarse
  grain scheduling
• Parallel task models moldable tasks, divisible
  load
• Some results
• Comparisons and coupling models IJ FCS 06
• Off-line improvement of performance ratio
• 3/2-approximation SIAM J.Comp 07 instead of 2
  Turekal by strip-packing
• (3?5) for moldable tasks on a grid of clusters
  Europar06
• On-line decrease of control overhead
   work-first principle  Cilk
• Extension to general distributed data-flow
  computations ICTTA06, ICCS07

Task
9
1. Scheduling

• Objective B Design of multi-objective scheduling
  with provable guarantees.
• Simultaneous approximation for each objective
• Approximated solutions of Pareto optimal
  solutions
• Makespan/ReliabilitySPAA07 - Makespan/Memory
  IPDPS08

• Generic ?-Relaxation scheme Shmoysal.
• Makespan/Minsum WEA05

To include a smart algorithm inside a recursive
doubling (eg. for Makespan)
(eg. for Minsum)
For moldable tasks yields a bi-approximation
with arbitrary ratio between Cmax and Minsum
WEA05
t
16
0
2
4
8
10
2. Interfaces for coordination

• Objective provably efficient control at
  runtime of the coupling of
  components with various synchronizations
  constraints.
• Kaapi middleware
• Provable performances
• Efficient local serialization work-first
  principle, zero-copy J. CLSS07, ICCS07
• Scheduling
• coarse-grain graph partitioning
  work-stealing
• Fault-tolerance protocols, from scheduling
  properties
• coordinated protocol ICTTA06 original TIC
  protocol EIT05, TDSC08
• Positioning
• Multi-processors/multi-core architectures Intel
  TBB, Cilk
• Grid / global platforms Tolerate failure and
  falsification Satin (FT)

1 struct sum 2 void operator()(Shared_r
lt int gt a, 3 Shared_r lt
int gt b, 4 Shared_w lt int
gt r ) 5 r.write(a.read() b.read())
6 7 8 struct fib 9 void
operator()(int n, Shared_wltintgt r) 10 if
(n lt2) r.write( n ) 11 else 12
int r1, r2 13 Forklt fib gt() ( n-1, r1 )
14 Forklt fib gt() ( n-2, r2 ) 15
Forklt sum gt() ( r1, r2, r ) 16 17
18
Local stack
runtime
Distributed nestedmacrodataflow graph
11
2. Kaapi Support and transfert

• Distributed implementation of CAPE-Open standard
  for process engineering computations IFP
• Cluster implementation of compliant runtime
  RSI/Indiss-RT

• Quadratic assignment ANR CHOC
• Finite element computations ANR DISCOGRID
• Cryptographic S-Box selection ANR SAFESCALE
• Probabilistic inference engine ProBayes

12
3. Adaptive algorithms
Objective To design and analyze algorithms that
may obliviously adapt their execution under the
control of the scheduling

Sequential algorithm
Parallel algo 1 P2
Parallel algo 2 P100
Parallel algo k P8
Which one to select?
13
3. Adaptive algorithms

• Heterogeneous resources, variable speeds
  work-stealing to obliviously self-tune
  granularity
• But work Wp increases when depth Dp decreases
• multi-objective problem
• Adaptive recursive coupling of algorithms
  Europar06, PASCO07, PDP08
• Relaxation sequential / parallel work-stealing

• Minimize both the work Wp and the depth Dp

14
3. Adaptive algorithms

• Cacheprocessor oblivious stream computations
  PDP07
• AWS adaptive work-stealing for MPSoCs
• Use case HDTV on MPSoCs ST Microelectronics film
  grain tech.

MPSoC
Application description potential
parallelism AWS api
Architecture description SPIRIT / IP-XACT
Simulator

• Near optimal experimental results PDP08

15
3. Adaptive algorithms

• Adaptive 3D-vision VR07
• Realtime constraint 30 frames per sec
• Adaptive heterogeneous coupling with Kaapi
  CPUGPU EGPV07

Maximum precision
Level of details
1 .. 16 CPUs
16
4. Interactivity

• Motivation parallelism for interactive
  applications
• Challenging application multi-cameras,
  multi-cpus, multi-GPUs, multi-display
• Grimage platform 2004
• Positioning other platforms
• Blue-C, ETH Zurich, 2005 ,Tele-Immersion_at_UCBer
  keley 2005
• Specificity collaboration with
• EVASION(realtime physics simulation)
• PERCEPTION (computer vision)

• - 30 nodes cluster
• - 15 cameras
• - 16 projectors

17
4. Interactivity

• Middleware
  dedicated to
  interactive applications
• Distributed components, moldable
• Parallel code coupling
• Static coarse grain mapping

HDTV player on 12 Mpixels display wall (16
projectors) CPUs GPUs
18
Summary of 2005-2007
AWS

• Multi-objective Adaptive Performance
• Applications are time-consuming but essential to
  validate scientific approach

19
Some facts

• Publications
• Contracts
• Softwares
• Kaapi, FlowVR, Taktuk, AWS

• 127 in 3 years , 19 rank 1 - 17 Int. Journal
  (SIAM J.Comp, IEEE TC, TPDS, TDSC, EJOR, FCS,
  )
• - 59 Int. Conf (SPAA, IPDPS, CCGrid VR,
  VIS, Europar, ICCS, Siggraph)

• Industry partners STM, IFP, CEA, Bull,
  C-S, DCN,
• 2 ARC, 5 ANRs
• 1 pole MINALOGIC
• 2 Europe, 1 Ass. team

K
20
Highlights

• 1st prize Plugtest Nov. 2007 Nqueens challenge
  
• SIGGRAPH Aug. 2007 Emerging Technologies Demo
• Valorization start-up (Sep. 2007)
• co-founded by former PhD C. Menier joined MOAIS
  / PERCEPTION
• transfer parallel 3D modeling

• Dec. 2006 special Jury prize
• Nov. 2007 1st prize
• Nqueens(23) in 2107s with 3654 cores

4000 visitors
21
Research directions 2008-20121/3

• To push the interactions to large scale
• Heterogeneous computing
• Complex memory hierarchy
• Provable performances vs adversary
• Game theory

22
Research directions 2008-20122/3

• Scheduling multi-objective
• Large systems, many users, various objectives
  equity / fairness
• Extra global objective to non-cooperative
  strategies
• Coordination interface gt Runtime for HIPC on
  demand
• Work-stealing based runtime extended to complex
  memory hierarchy
• Dependable computing on global computing
  platforms

23
Research directions 3/3

• Adaptive algorithms
• Large data sets, out-of-core issues
• Framework / high level library
• High performance interactive computing
• Interactive resolution of complex problem
  (scheduling)
• Grimage explore new 3D interactions PERCEPTION
• Parallelism for adaptive interactive performance
• EVASION, ALCOVE
• Kaapi partitioningwork-stealing to balance
  load between heterogeneous resources (CPUs / GPUs
  )

24
Summary

• To provide parallel programming schemes,
  interfaces and tools for high performance
  interactive computing that enable to achieve
  provable performances on distributed parallel
  architectures, from multi-processors
  system-on-chip to lightweight grids and global
  computing platforms.

SIGGRAPH07 MOAIS - PERCEPTION - EVASION
25
Former members

• 13 PhDs defended in 2005-2007
• Now 2 at INRIA Alcove, Cepage 7
  in university Reims, IKI Iran, Luxembourg,
  Vannes, Damascus, Warsaw, Colima 1
  in Postdoc Iowa SU 1 Start-up
  co-founder 4DViews 2 in industry IFP,
  Amadeus
• 2 postdocs
• Now Univ. Paris 6, Petrobraz
• 1 long term visit
• Axel Krings, Idaho State Univ
• 3 engineers
• Now INRIA/PARIS, industry