Title: Energy-Aware Modeling and Scheduling of Real-Time Tasks for Dynamic Voltage Scaling
1Energy-Aware Modeling and Scheduling of Real-Time
Tasks for Dynamic Voltage Scaling
- Xiliang Zhong and Cheng-Zhong Xu
- Dept. of Electrical Computer Engg.
- Wayne State University
- Detroit, Michigan
- http//www.cic.eng.wayne.edu
2Outline
- Introduction and Related Work
- A Filtering Model for DVS
- Time-invariant Scaling
- Time-variant Scaling
- Statistical Deadline Guarantee
- Evaluation
- Conclusion
3Motivation
- Mobile/Embedded devices power critical
- Energy-Performance tradeoff
- Processor speed designed for peak performance
- Slowdown the processor when not fully utilized
(DVS) - Challenges
- Maximize energy saving while providing deadline
guarantee - Real-time tasks could be periodic/aperiodic w/
highly variable execution time - Aperiodic tasks have irregular release times,
which calls for online decision making
4Related Work
- Intensive studies for periodical tasks
- Algorithms for aperiodic tasks
- Offline (Yao et al95, Quan Hu01)
- Online all timing information known only after
job releases - Soft real-time improve responsiveness (Aydin
Yang04) - Occasionally uncontrollable deadline misses
(Sinha Chakrabarty01) - Hard real-time w/complex admission control (Hong
et al 98) - Maximize energy saving w/ frequency scaling(Qadi
et al 03, DVSST) - On-line slack management for a general input (Lee
Shin 04,OLDVS) - Objectives of this paper
- Hard/statistical deadline guarantee for general
input w/o assumptions of task periodicity - Unified, online solutions for both WCET based
scheduling and slack management
5Task Model
- Independent tasks, preemptive w/ dynamic
priorities - Job releases (requests) to system are
characterized by a compound process in a discrete
time domain - wi (t) is the size (WCET) of ith jobs arrived
during time t-1,t) - n(t) stands for number of jobs arrived, each
w/deadline td
Input arrivals
w1(1) w1(2) w2(3)
time
2
1
0
6System Model
- Processor Model
- Support a continuous range of speed levels
- Energy Model
- t scheduling time slot, f(t) speed at time t,
t1) - l(t) load, cycle allocated to all jobs during
t, t1) - P(l(t)) power as a function of load
- E(S) energy consumed according to a schedule S
7A Filtering Model of Speed Scaling
- Allocation function denotes the cycles
allocated to one job wi(t) during t, t1) - Decomposition of allocation function
- g(), the impact of job sizes (WCETs) on
scheduling - h(), scaling function
- s(), the load feedback to scheduling
8A Filtering Model (cont.)
- Load Function l(t) is a sum of allocation to all
jobs
- Each job should be finished in td time
g(wi(t))wi(t)
- Non-adaptive to load s(l(t)) 1
9A Filtering Model (cont.)
- The load function becomes a convolution of
compounded input request process and scaling
function, - Scaling function h(t) Portion of resource
allocated at each scheduling epoch from the
arrival time ts to finish time tstd - Design of scaling algorithm in a fitlerng system
10Time-Invariant Scheduling
- Treat h(t) as a time-invariant scaling function
- The optimal policy is to find an allocation
where
- The optimality is determined by the covariance
matrix ? of the input process w(t) in the order
of deadline td - The optimization has a unique, closed form
solution
11Example Solutions with Different Input
- Two multimedia traffic patterns (Krunz00)
- Shifted Exponential Scene-length Distribution
(ACFExp) - Subgeometric scene-length distribution
(ACFSubgeo) - Fractional Gaussian Noise (FGN) process with
Hurst para. H0.89 - Simpsons MPEG Video Trace of 20,000 frames
Auto-Correlations of Traffic
12Example Solution (td10)
- Higher degree of input autocorrelation has a more
convexed scaling function - The uniform distributed allocation is a
generalization of several existing algorithms for - Periodic tasks
- Sporadic tasks
- Aperiodic tasks
13Time-Variant Scaling
- Energy consumption can be reduced if the scaling
function h(t) is adaptive in response to change
of input load
- Make td runnable queues. Jobs with deadline j are
put to queue j
14Time-Variant Scaling
- Minimize energy consumption is to
subject to
where qj(t) is the backlog of queue j at time t.
- The optimization has a unique solution
Resource cap of queue j at time t
Committed resource for jobs in queue j at time t
15Illustration
- Determine cap of queue 5 at time 0 S5(0)
load
- First determine current committed resource
- Distribute the job as late as possible
- The job is distributed to early slots as its size
increases
0 1 2 3 4 5 time
slot
16Example Solution for a Sporadic Task
Input J(WCET) J1(1) released at 0, 5, J2(2) at
1, 7, J3 (1) at 3, 9. Deadline of all jobs 4.
Ji,j jth instance of task i
1. Schduling using EDF w/o scaling
2. Schduling using the Time Variant Scaling
Using a square energy function 35 more energy
saving compared to EDF. 8 to DVSST
17Statistical Deadline Guarantee
- Worst case scenario schedulability test
- Conservative
- pi minimum interarrival
- Statistical guarantee
- Overload probability vprob(l(t) gt fmax)
cumulative probability
1
v
F(x)
fmax
worst case f
18Statistical Deadline Guarantee (cont.)
- Load tail distribution
- A general bound w/ load mean and variance
- Tight bounds based on load distribution
- Exact output distribution if input distribution
known - Estimate output distribution using a histogram
fmax
19Evaluation
- Objectives
- Effectiveness in energy savings
- Effectiveness of the deadline miss bound
- Scheduling based on WCET
- No-DVS run jobs with the maximum speed.
- Offline Offline optimal algorithm of Yao95 et
a. - DVSST On-line algorithm for sporadic tasks
QadiRTSS03 et al. - TimeInvar Time-invariant voltage scaling.
- TimeVar Time-variant voltage scaling.
- On-line slack management
- DVSSTCC (Cycle-conserving EDF) Worst case
schedule using DVSST with the reclaiming
algorithm of Pillai and Shin (SOSP01). - TimeVarOLDVS The time-variant voltage scaling
and the reclaiming algorithm of Lee and
ShinRTSS2004. - TimeVarTimeVar A unified solution.
20Energy Savings
- Energy consumption with the Robotic Highway
Safety Marker application A scenario in which
robot keeps moving
DVSST
Offline
TimeVariant
- TimeVar is energy-efficient, close to Offline
(5) 7-11 better than DVSST
21Energy Savings w/ Workload Variation
- tasks30
- Interarrival exp(50 ms)
- WCET n(100, 10)K
- Workload variation characterized by actual
execution time over worst case (BCET/WCET)
TimeVariant adapts with workload variation
effectively
22Computation Speed Configuration
Required speed (MHz)
- Target deadline guarantee 99
Mean Interarrival time (ms)
- Computation requirement based on a general bound
is better than worst case with mean interarrivals
gt 60 ms - Tight bounds reduce the computation speed in half
as interarrivals gt 40 ms
23Statistical Deadline Guarantee
- No deadline misses under bound derived based on a
general input 100MHz - Statistics of TimeVar/TimeInvar under a tight
bound 40MHz - Overload handling reject new jobs or serve
unfinished jobs in a best-effort mode - Target deadline guarantee 99
Scheduling TimeInvariant TimeInvariant TimeVariant TimeVariant
Scheduling Reject Besteffort Reject Besteffort
Load mean (106) 21.6 21.7 21.6 21.74
Load var 166.7 168.9 141.3 142.8
Time mean 10.1 10.09 10.4 10.07
Time variance 8.7 8.4 8.4 8.8
Overload/Deadline misses 0.63 0.63 0.61 0.61
Deadline miss rate is effectively bounded
24Conclusion
- Voltage/Speed scaling for a general task model
- A Filtering Model for DVS
- Two online policies to minimize energy usage
- Time-invariant A generalization of several
existing approaches - Time-variant Optimal in the sense it is online
w/o future task timing information. Also
effective for on-line slack management - Statistical deadline guarantee based on
computation speed configuration. - Future work
- System-wide energy savings, e.g., wireless
communication and its interaction with CPU
25Energy-Aware Modeling and Scheduling of Real-Time
Tasks for Dynamic Voltage Scaling
Thank you!