energy/reliability trade-offs in fault-tolerant event-triggered distributed embedded systems
DESCRIPTION
Energy/Reliability Trade-offs in Fault-Tolerant Event-Triggered Distributed Embedded Systems. Junhe Gan, Flavius Gruian, Paul Pop, Jan Madsen. Motivation. The impact of power supply scaling on fault-tolerance (MTTF) and power dissipation - PowerPoint PPT PresentationTRANSCRIPT
Energy/Reliability Trade-offs in Fault-Tolerant Event-Triggered Distributed Embedded Systems
Junhe Gan, Flavius Gruian, Paul Pop, Jan Madsen
2
MotivationThe impact of power supply scaling
on fault-tolerance (MTTF) and power dissipation
Maheshwari, A. et al., Trading off transient fault tolerance and power consumption in deep submicron (DSM) VLSI circuits , VLSI Systems, pp. 299-311, 12(3), 2004.
3
Outline
Architecture, Application and Energy Models Reliability model Energy/reliability trade-off model Schedulability analysis
Problem Formulation Example
Optimization Strategy TABU search-based algorithm
Experiment ResultsConclusion and Contributions
4
Architecture Model
A set of heterogeneous Processing Elementsinterconnected by a communication channel.
Each PE has a set of operating mode Each operating mode, we know ( , , )j j jN N N
i i if v p
5
Reliability Model
t1 t1 t1
Re-execution:Transient faults
t1
t1
t1
Replication:Transient faults /Permanent faults
N1
N2
N3
K = 2
1
1
1 (1 )
1 1
i
ci
reex ki
krepi i
i
A iP A
R e
R R
R R
R R
6
Energy and Application ModelApplication Model:
A set of periodic tasks for each task, we know
Unique priority Critical/non-critical task
Energy Model:
j
i j
N i imap on N
Energy Power Time
E p t O
( , , )jN
i i iC T D
Sum of mode-switching
overheads
7
Energy/reliability Trade-off ModelThe average failure rate of a system increases exponentially
when the supply voltage and the operating frequency decreases
0( , ) 10 VF V F
8
Schedulability AnalysisFixed-Priority
Preemptive Scheduling
Response Time Analysis
Degree of Schedulability
( )j i
ii i j
hp j
rr c c
T
1 1
2 1
max(0, ) 0
( ) 0
m
i i iiA m
i i ii
d r D if dS
d r D if d
1 2 3priority priority priority
9
Problem FormulationGiven:
Application Architecture Reliability Goal Tolerate k-faults for each task
We are interested to determine: the mapping for allocating each task the operating mode for executing each task
Such that: all tasks meet their timing requirements the application reliability meets the given reliability goal the energy consumption of system is minimized
10
Motivational Example
11
Optimization Strategy
NP-hard problem: ways of allocating tasks to PEs. ways of executing tasks at operating modes of an PE.
Search the optimal solution by heuristic algorithm.
Minimize the cost function:( ) max(0, ) max(0, )S R g s r sCost S E W R R W r
mnmk m
m nk
Energy Consumption Reliability
Goal
TimingRequirement
12
Searching Loop:
Tabu Search-based Algorithm
Save Best Solution (if it is)
Select New Solution
First attempt: An improved solution Otherwise: Randomly select
a non-improved & not-tabu solution
Operating Mode (L-moves)Change a task’s operating mode
Mapping (M-moves)
Move a task from an PE to another PE Or Swap two tasks between PEs
Design Transformations
Use and Maintenance of
Tabu-list
13
Experimental Results
14
Experimental Results
15
Conclusion and ContributionsConclusion:
Contributions: Energy/reliability model Combine deciding mapping and operating mode assignment Optimization algorithms for this energy/reliability trade-offs problem
It is possible to eliminate the negative impact of energy minimization on reliability
by carefully considering the energy/reliability trade-offs during optimization.