agenda
DESCRIPTION
ProActive Parallel Suite: Multi-Cores to Clouds to Autonomicity. D. Caromel, et al. 1. Background: OASIS, ActiveEon 2. ProActive Overview 3. Programming (Components: GCM Standard) 4. Optimizing 5. Scheduling + Resourcing 6 . SOA, SLA and QoS. Agenda. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/1.jpg)
Agenda1. Background: OASIS, ActiveEon2. ProActive Overview 3. Programming (Components: GCM Standard)4. Optimizing5. Scheduling + Resourcing6. SOA, SLA and QoS
D. Caromel, et al.
ProActive Parallel Suite:Multi-Cores to Clouds to Autonomicity
Parallelism+Distribution with Strong Model: Speed & Safety
![Page 2: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/2.jpg)
Key Objectives Parallel Programming Model and Tools
desesperatly needed for the masses for new architectures (Multi-cores)
As Effective as possible: EfficientHowever Programmer Productivity is first KSF
For both Multi-cores and DistributedActually the way around
Handling of ``Large-scale’’ (Grid, Clouds)
![Page 3: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/3.jpg)
33
1. Background1. Background
![Page 4: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/4.jpg)
44
OASIS Team & INRIA
A joint team, Now about 35 persons 2004: First ProActive User Group 2009, April: ProActive 4.1, Distributed & Parallel:
From Multi-cores to Enterprise GRIDs
![Page 5: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/5.jpg)
55
OASIS Team Composition (35) Researchers (5):
D. Caromel (UNSA, Det. INRIA) E. Madelaine (INRIA) F. Baude (UNSA) F. Huet (UNSA) L. Henrio (CNRS)
PhDs (11): Antonio Cansado (INRIA, Conicyt) Brian Amedro (SCS-Agos) Cristian Ruz (INRIA, Conicyt) Elton Mathias (INRIA-Cordi) Imen Filali (SCS-Agos / FP7
SOA4All) Marcela Rivera (INRIA, Conicyt) Muhammad Khan (STIC-Asia) Paul Naoumenko (INRIA/Région
PACA) Viet Dung Doan (FP6 Bionets) Virginie Contes (SOA4ALL) Guilherme Pezzi (AGOS, CIFRE
SCP)
+ Visitors + Interns
PostDoc (1): Regis Gascon (INRIA)
Engineers (10): Elaine Isnard (AGOS) Fabien Viale (ANR OMD2, Renault ) Franca Perrina (AGOS) Germain Sigety (INRIA) Yu Feng (ETSI, FP6 EchoGrid) Bastien Sauvan (ADT Galaxy) Florin-Alexandru.Bratu (INRIA CPER) Igor Smirnov (Microsoft) Fabrice Fontenoy (AGOS) Open position (Thales)
Trainee (2): Etienne Vallette d’Osia (Master 2 ISI) Laurent Vanni (Master 2 ISI)
Assistants (2): Patricia Maleyran (INRIA) Sandra Devauchelle (I3S)Located in Sophia Antipolis, between Nice and Cannes,
Visitors and Students Welcome!
![Page 6: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/6.jpg)
66
Co-developing, Support for ProActive Parallel Suite Worldwide Customers: Fr, UK, Boston USA
Startup Company Born of INRIA
![Page 7: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/7.jpg)
Multi-Cores
7
![Page 8: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/8.jpg)
88
Symetrical Multi-Core: 8-ways Niagara II
8 cores 4 Native
threads per core
Linux see 32 cores!
![Page 9: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/9.jpg)
99
Sun 16-core Rock: Fall 2009
16 cores 4 native threads per core
64 “Cores” or “Native Threads” at OS level
![Page 10: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/10.jpg)
1010
Intel 8-cores, 16-thread Nehalem-based Xeon processor confirmed (Feb. 2009)
Highly NUMA
Not an SMP:
L1, L2, thenL3 attachedto a givencore
![Page 11: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/11.jpg)
Multi-Cores A Few Key Points
Not Shared Memory (NUMA) Moore’s Law rephrased:
Nb. of Cores double every 18 to 24 months Key expected Milestones: Cores per Chips (OTS)
2010: 32 to 64 2012: 64 to 128 2014: 128 to 256
1 Million Cores Parallel Machines in 2012 100 M cores coming in 2020 Multi-Cores are NUMA, and turning Heterogeneous (GPU) They are turning into SoC with NoC: NOT SMP!
![Page 12: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/12.jpg)
12
2. OverviewProActive Parallel Suite
![Page 13: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/13.jpg)
1313
2. Programming OptimizingParallel Acceleration Toolkit in Java:
- Java Parallelism + Legacy-Code Wrapping and Control - Scheduling and Resource Manager
Multi-Core + Distributed
Open Source Used in production by industry
![Page 14: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/14.jpg)
1414
OW2: Object Web + Orient Ware
![Page 15: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/15.jpg)
1515
![Page 16: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/16.jpg)
1616
ProActive Contributors
![Page 17: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/17.jpg)
1717
3. ProActive Programming: Active Objects
![Page 18: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/18.jpg)
1818
![Page 19: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/19.jpg)
1919
![Page 20: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/20.jpg)
202020
A
ProActive : Active objects
Proxy
Java Object
A ag = newActive (“A”, […], VirtualNode)V v1 = ag.foo (param);V v2 = ag.bar (param);...v1.bar(); //Wait-By-Necessity
V
Wait-By-Necessity is a
Dataflow Synchronization
JVM
A
JVM
Active Object
Future Object Request
Req. Queue
Thread
v1v2 ag
WBN!
![Page 21: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/21.jpg)
212121
Standard system at Runtime: No Sharing
NoC: Network On ChipProofs of Determinism
![Page 22: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/22.jpg)
Key Point: Software Evolution
Distributed To Multicores Multi-Cores: 32 (2010) to 64 to 128 to 256 (2014)Shift the execution from several multi-cores executingthe same application simultaneously to a single, larger multi-core chip. An application requiring 128 cores to correctly execute, can be executed in 2012 on four 32 cores, and seamlessly executed in 2016 on a single 128-core chips
Smooth evolutivity of applications:Distributed and Multi-core Platforms
![Page 23: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/23.jpg)
232323
Standard system at Runtime: No Sharing
NoC: Network On ChipProofs of Determinism
![Page 24: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/24.jpg)
Key Point: Locality will more than ever be
Fundamental
Let the programmer control it
No global shared memory
![Page 25: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/25.jpg)
2525
TYPED ASYNCHRONOUS GROUPS
![Page 26: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/26.jpg)
262626
A
Creating AO and Groups
Typed Group Java or Active Object
A ag = newActiveGroup (“A”, […], VirtualNode)V v = ag.foo(param);...v.bar(); //Wait-by-necessity
V
Group, Type, and Asynchrony are crucial for Composition
JVM
![Page 27: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/27.jpg)
272727
Broadcast and Scatter
JVM
JVM
JVM
JVM
agcg
ag.bar(cg); // broadcast cgProActive.setScatterGroup(cg);ag.bar(cg); // scatter cg
c1 c2 c3c1 c2 c3
c1 c2 c3c1 c2 c3c1 c2 c3
c1 c2 c3
s
c1 c2 c3
s
Broadcast is the default behavior Use a group as parameter, Scattered depends on rankings
![Page 28: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/28.jpg)
282828
Dynamic Dispatch Group
JVM
JVM
JVM
JVM
agcg
c1c2c3
c4c5
c6c7
c8c0c9c1
c2c3
c4c5
c6c7
c8c0c9
c1c2c3
c4c5
c6c7
c8c0c9
Slowest
Fastest
ag.bar(cg);
![Page 29: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/29.jpg)
Abstractions for Parallelism
The right Tool to do the Task right
![Page 30: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/30.jpg)
3030
ProActive Parallel Suite
Workflows in Java Master/Workers SPMD Components …
![Page 31: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/31.jpg)
31
Components: GCM Standard
31
![Page 32: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/32.jpg)
3232
GridCOMP Partners
![Page 33: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/33.jpg)
3333
GCM Standardization
Grid Component Model
Overall, the standardization is supported by industrials:
BT, FT-Orange, Nokia-Siemens, NEC,Telefonica, Alcatel-Lucent, Huawei …
![Page 34: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/34.jpg)
343434
Objects to Distributed Components
Typed Group Java or Active Object
V
AExample ofcomponentinstance
JVM
Truly Distributed
Components
IoC:InversionOf Control(set in XML)
![Page 35: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/35.jpg)
35
GCM Components
Scopes and Objectives:Grid Codes that Compose and DeployNo programming, No Scripting, … No Pain
Innovation:Abstract DeploymentComposite Components Multicast and GatherCast
MultiCast GatherCast
![Page 36: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/36.jpg)
3636
GCM StandardizationFractal Based Grid Component Model
4 Standards:
1. GCM Interoperability Deployment2. GCM Application Description 3. GCM Fractal ADL4. GCM Management API
![Page 37: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/37.jpg)
Key Points about
Parallel Components
Parallelism is captured at the Module interface Identical to Typing for functional aspectsComposition, parallel word, becomes possible Configuration of the Parallel aspects
![Page 38: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/38.jpg)
3838
![Page 39: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/39.jpg)
3939
4. Optimizing
![Page 40: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/40.jpg)
4040
![Page 41: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/41.jpg)
4141
IC2D
![Page 42: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/42.jpg)
4242
IC2D
![Page 43: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/43.jpg)
4343
ChartIt
![Page 44: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/44.jpg)
4444
Pies for Analysis and Optimization
![Page 45: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/45.jpg)
45
Video 1: IC2D OptimizingMonitoring, Debugging, Optimizing
![Page 46: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/46.jpg)
4646
5. Scheduling & Resourcing
![Page 47: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/47.jpg)
4747
![Page 48: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/48.jpg)
48
ProActive Scheduling Big Picture
RESOURCES
Multi-platform Graphical Client (RCP)
File-based or LDAP authentication Static Workflow Job Scheduling, Native and
Java tasks, Retry on Error, Priority Policy, Configuration Scripts,…
Dynamic and Static node sources, Resource Selection by script, Monitoring and Control GUI,…
ProActive Deployment capabilities: Desktops, Clusters, Clouds,…
ProActiveScheduler
ProActiveResource Manager
![Page 49: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/49.jpg)
4949
Scheduler: User Interface
![Page 50: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/50.jpg)
50
Job
Another Example : Picture Denoising
Split
Denoise DenoiseDenoiseDenoise
Merge
• with selection on native executable availability (ImageMagik, GREYstoration)• Multi-platform selection and command generation
• with file transfer in pre/post scripts
![Page 51: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/51.jpg)
51
Video 2:Scheduler, Resource Manager
![Page 52: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/52.jpg)
5252
Clusters to Grids to Clouds:
e.g. on Amazon EC2
![Page 53: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/53.jpg)
Node source Usecase : Configuration for external cloud with EC2
ProActiveSchedulerProActive
Resource Manager
Dedicated resources
LSF
Static Policy
Amazon EC2
EC2
Dynamic Workload Policy
Desktops
SSH
Timing Policy 12/24
![Page 54: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/54.jpg)
54
Video 3:Provisioning Resources from Clouds
![Page 55: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/55.jpg)
5555
6. SOA, SLA and QoS
![Page 56: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/56.jpg)
5656
AGOS: Grid Architecture for SOA
AGOS Solutions
Building a Platform for Agile SOA with Grid
In Open Source with Professional Support
![Page 57: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/57.jpg)
5757
AGOS Generic Architecture for Autonomic SOA with GRIDs & Clouds
OS, HW
OS Virtualization Grid Utility interface
ESBEnterprise Service Bus
SCA Service Component Architecture
Resource Manager
Task & Services Scheduling
Parallel ProgrammingSPMD, workflow
Agent, Master/WorkerFork and Join
In memory db cache(JSR / JPI / javaspaces) SOA BPEL Exec
Repository, Registry, Orchestration
SOA MonitoringReporting, Notifications,
alarms
Business Intelligence BI Monitoring
SLM SLM
SLM
SLM
SLMSLM
SLMSLM
SLM SLM
Ser
vice
Lev
el M
anag
emen
t
![Page 58: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/58.jpg)
58
Conclusion
58
![Page 59: Agenda](https://reader036.vdocument.in/reader036/viewer/2022081604/5681655c550346895dd7dc19/html5/thumbnails/59.jpg)
5959
Conclusion:
An Acceleration Toolkit :Concurrency+ParallelismMulti-Core+Distributed