Dynamic adaptation of parallel codes Toward self-adaptable components for the Grid

1
Dynamic adaptation of parallel codesToward
self-adaptable components for the Grid

• Françoise André, Jérémy Buisson Jean-Louis
  Pazat
• IRISA / INSA de Rennes / Université de Rennes 1

2
Our view of the Grid
Cluster resource
Cluster resource
Application
WAN
Our point of interest
Cluster resource
3
Our view of the Grid
Cluster resource
Processor resources
Network resource

4
Our view of the Grid

• Environment that is
• Parallel
• Grid is built up from parallel machines
• Dynamic
• Resource allocation may change dynamically
• Distributed
• Resources are distributed over a network
• Resources are in different administration domains
• Need for a new programming technique
• Parallel self-adaptable distributed components

5
Related works

• Parallel and distributed components / objects
  exist
• Example GridCCM, PARDIS
• Self-adaptable components exist
• Example ACEEL, DART
• But no parallel and self-adaptable distributed
  component

6
Principles of parallel components

• Encapsulation of a parallel code
• Collaboration of several communicating processes
• Goal allow to easily couple parallel codes

7
Principles of dynamic adaptation

• Modification of the executed code
• Reflexive programming
• Goal better fit to allocated resources

Execution flow
1. Event
2. Reaction
8
Dynamic adaptation

• Three key questions
• When should the component adapt?
• How should the component be modified?
• Where can the reaction be executed?

9
Dynamic adaptation

• When should the component adapt?
• Upon reception of an event from a monitor
• According to the policy

Monitor
Decider
Notifies of events
Interprets
Adaptation policy
10
Dynamic adaptation

• How should the component be modified?
• Executing special code
• Following directives of the policy

Coordinator
Requests execution of reactions
Decider
Executes
Interprets
Reaction
Adaptation policy
11
Dynamic adaptation

• Where can the reaction be executed?
• At the next adaptation point
• Approximated prediction of the next point
• Based on control flow graph

Behavior 1
Behavior 2
Reaction
Not an adaptation point
An adaptation point
12
Dynamic adaptation
Coordinator
Requests execution of reactions
Monitor
Decider
Executes
Notifies of events
Platform
Component
Interprets
Reaction
Adaptation policy
Modifies
Behavior
13
Mixing parallelism and adaptation
Parallel coordinator
Requests execution of reactions
Monitor
Decider
Executes
Notifies of events
Platform
Component
Interprets
Parallel reaction
Adaptation policy
Modifies
Parallel behavior
Parallel behavior
Parallel behavior
14
Mixing parallelism and adaptation

• Introduction of global adaptation points
• All the processes at the same state
• Need to coordinate all the processes
• Example SPMD code
• Adaptation point between each phase

Local adaptation point
Global adaptation points
Not global adaptation points
15
Mixing parallelism and adaptation

• Need for a distributed algorithm for the parallel
  coordinator
• Only consider globally reachable points
• In the future of all the processes
• Make an agreement of all the processes
• Choose the same point for all the processes

16
Mixing parallelism and adaptation

• Need to control the non-determinism
• Due to parallelism
• Dynamically insert synchronization statements
• Due to unpredictable conditional instructions
• Force the result of the conditions if possible
• Example insertion of empty iterations in loops
• Otherwise postpone the decision-making

17
Experiment

• Experiment
• Iterative SPMD code
• Adaptation points between each iteration
• Increase of the number of processors
• Results
• Negligible time in adaptation points
• Gain thanks to the adaptation
• Expected to scale well

Adaptation
18
Related domains
Man-Machine Interface

• Computation steering
• Notions equivalent to global adaptation points
• Need to execute some special code at the next
  special point
• Particular use of adaptation mechanisms
• User interface instead of monitors

Parallel coordinator
Requests execution of reactions
Monitor
Decider
Executes
Notifies of events
Platform
Component
Interprets
Parallel reaction
Adaptation policy
Modifies
Parallel behavior
Parallel behavior
Parallel behavior
19
Related domains

• Fault tolerance
• Consider dynamic environment
• Need for a global consistent state
• In the past for fault tolerance
• In the future for dynamic adaptation
• Relation to dynamic adaptation
• An application?
• A complementary feature?

20
Work done

• Design of the overall architecture
• Identification of functional boxes
• Distributed algorithm for the coordinator
• Automated instrumentation by static behavioral
  reification
• Simple negotiation protocol
• Demonstration prototype
• Ad-hoc mechanisms
• Proof of concept

21
Future work

• Generalizing the approach
• Generic definition of global adaptation points
• Limits of the same state definition
• Case of non-SPMD codes
• Expression of the adaptation policy
• Limits of explicit event-based rules
• Need for more sophisticated (intelligent?)
  policies
• Smoothing measures of resource availability
• Balancing instabilities

22
Future work

• Collaborative adaptation of components
• Control side-effects
• Avoid adaptation cycles
• Common policy at the level of
• A group of components
• A composite
• The whole application
• Consider full Grid applications
• Not only their components

23
Dynamic adaptation of parallel codesToward
self-adaptable components for the Grid