This document discusses energy efficient real-time systems. It introduces real-time systems and divides them into hard and soft categories. Hard real-time systems have consequences for missing deadlines while soft systems do not. The document then discusses embedded systems and the need for energy efficiency in real-time systems. It presents scheduling and dynamic voltage scaling techniques to improve energy efficiency. Finally, it proposes a dynamic scheduling algorithm to minimize energy consumption using an energy beneficial break even time approach.

1. Presented By:
Pragya Arya
Btech(cs) 3rd year
Banasthali University

2. INTRODUCTION
Energy power efficient real time systems
are systems which are power efficient i.e.
uses less energy for computation and
processing in the specified time which
basically lead to light weighted devices
having longer battery life.

3. REAL TIME SYSTEMS
 A real-time system is any information processing
system which has to respond to externally
generated input stimuli within a finite and
specified period.
 Mainly Real time systems are divided in two
categories
 Hard real time Systems
 Soft real time systems

4. HARD REAL TIME SYSTEMS
 An overrun in response time leads to potential
loss of life and/or big financial damage
 Many of these systems are considered to be
safety critical.
 Examples: Pacemakers , Airplane
control systems.

5. SOFT REAL TIME SYSTEMS
 Deadline overruns are tolerable, but not desired.
 There are no catastrophic consequences of
missing one or more deadlines.
 Example: live video
streaming

6. EMBEDDED SYSTEMS
 An embedded system is a computer system with
a dedicated function within a larger mechanical
or electrical system, often with a real time
constraint.

7. NEED OF ENERGY POWER EFFICIENCY
 Critical design issue in real-time systems,
especially in battery- operated systems.
 Smaller and lighter devices
 Less heat generation
 Better performance with unprecedented speed
 Reduce cost
 Reduce energy use
 Longer battery life.

8. POWER ISSUE CAN BE ADDRESSED IN
 Architecture Level
 System Level
 Application Level

9. SCHEDULING
 Scheduling is method by which threads , process,
or data flow are given access to system resources.
 This is usually done to load balance and share
system resources effectively.
 There are various energy efficient scheduling
algorithms.

10. DYNAMIC VOLTAGE SCALING
 Dynamic voltage scaling is a power
management technique in computer architecture.
 In which the voltage used in a component is
increased or decreased, depending upon
circumstances.
 Dynamic voltage scaling to decrease voltage is
known as undervolting. It is done in order to
conserve power.
 Dynamic voltage scaling to increase voltage is
known as overvolting. It is done in order to
increase computer performance.

11. SYSTEM MODELS
 The real time system we are interested in consist
of N independent periodic tasks, T = {τ1, τ2,⋅⋅⋅, τN}
scheduled according to Earliest Deadline First
(EDF).
 τi = (Ci , Di ,Pi), is characterized by its worst case
execution time Ci, deadline Di, and period Pi. We
assume Di<=Pi.
 The Jth job of task τi is represented with Ji j = (ri
j, ci j,di j),where ri j,ci j, and di j are the arrival
time, actual execution cycles, and absolute
deadline, respectively.

12.  Each task τi is associated with a subset of
peripheral devices Φi = {M0,M1, ...,Mk}.
 Φi can be further divided into shared part
Φi share and non-shared-part Φi nonshare.
 Each peripheral device can be in one of the two
states Active State or Shut down state.
 Energy overhead (Eo) andtime overhead (To) need
to be consumed to shut-down and later wake up
the processor or devices

13. Break even time is used to reflect
whether it is worthwhile to shut down the
processor /(peripheral devices) during idle
interval or not.
Ibe=max{Eo/P(idle),To}

14. DYNAMIC SCHEDULING ALGORITHM
 Energy beneficial break even time Ieb is
computed dynamically for each job according to
the run-time conditions and shut down the
processor/devices whenever necessary.
THEOREM :
 Given task set T = {τ1, τ2, ..., τN} with tasks
ordered by increasing value of Di.
 All task deadlines can be guaranteed if any job
Ji of task τi is procrastinated by no more than Δi
time units,
 where Δi (called the procrastination time of task τi)

15. COROLLARY 1:
 T – task set scheduled according to EDF
 up(Ji)- upcoming job set in which each job Jk has
arrival time rk > ri .
 If Ji is the only job in the ready queue, all jobs in
T can meet their deadlines if the starting
execution time of up(Ji) is delayed to tLS, where
tLS = min (rk +Δk) (1)
Jk∈up(Ji )

16. COROLLARY 2:
 T – task set scheduled according to EDF.
 Ji current job to be executed in any time t. hp(Ji)
is the upcoming job set in which each job Jl has
arrival time rl > t and deadline dl < di.
 Then all jobs in T can meet their deadlines if the
starting execution time of hp(Ji) is delayed to
˜tLS, where
˜tLS = min (rl +Δl) (2)
l∈hp(Ji)

17. SCHEDULING ALGORITHM
1: Input: The current job Ji and the current
time t(cur);
2: if Ji is the only job in the ready queue then
3:Let edi = min{di, tLS};
//tLS is computed with 1
4: Compute Ii(eb) based on Ji’s current feasible
interval [t(cur),edi];
5: let fi be the expected completion time of Ji under
S(crit) ;
6: if (edi − fi) ≥ Ii(eb) then

18. CONT..
7: Execute Ji with S(crit) and shut down the
processor and devices in Φi at fi and set up the
wake up timer to be(tLS − fi);
8: // critical speed strategy with shut-down
9: else
10: Let t(na) be the earliest arrival time for the
upcoming jobs;
11: if fi ≥ t(na) and fi ≤ di then
12: Execute Ji with S(crit) ; // critical speed strategy
without shutting down shared devices in Φishare

19. 13: else
14: Execute Ji with s′i = (ci×si)/(edi−(tcur)) non-preemptively
within [t(cur),edi]; // DVS strategy
stretching to edi;
Start Deadline Start Deadline
μProc.Speed Time
Idle time
represent
s wasted
energy
Lower speed,
Lower voltage,
Lower energy
Energy ~ Work • Speed
Work
Work

20. CONT..
15: end if
16: end if
17: else
18: Compute ˜tLS based on equation (2);
19: Execute Ji with max{si, S(crit)} non-preemptively
within [t(cur),˜tLS] and shut down
its non-shared devices Φi nonshare upon its
completion;
20: end if

21. μProc. Speed
S1 S2 S3 D2 D3 D 1
W1 W2
Time
Task runs faster
to meet timing
constraints
W3

22. APPLICATION AREAS
 Customer products:
 Dish washers
 Microwave ovens
 Cars:
 Airbag system
 Engine control
 Planes
 Military
 Weapons
 Satellites
 Robotics
 Protection & security
system
 Intruder Alarm
 Smoke/gas detection

23. LIMITATIONS
 In DVS systems however, the performance level
is reduced during periods of low utilization such
that the processor finishes each task “just in
time,” stretching each task to its deadline,
 The primary caveat of overvolting is increased
heat: the power dissipated by a circuit increases
with the square of the voltage applied, so even
small voltage increases significantly affect power.

24. CONCLUSION
 Dynamic Scheduling Algorithm is used to
minimize the system level energy consumption.
 Approach of energy beneficial break even time is
is used to enhance the computation.
 This approach can reduce energy consumption for
processor/devices significantly when compared
with other approaches.

25. REFERENCES
 http://www.cse.unsw.edu.au/~cs9242/08/lectures/
09-realtimex2.pdf - last seen 21/9/14
 http://airccse.org/journal/ijcsea/papers/2212ijcsea
16.pdf - last seen 24/9/14
 http://www.dateconference.com/proceedings/PAP
ERS/2011/DATE11/PDFFILES/IP1_08.PDF - last
seen 21/9/14
 http://www.ecs.umass.edu/ece/koren/architecture/
RtDVS/intro.htm - last seen 28/9/14.