Round Robin Scheduling in Embedded Systems

2. SCHEDULING AND ITS TYPES
 Scheduling is a method by which threads, processes or data flows are given access to system
resources.
 The need for scheduling algorithm arises from the requirement for most modern systems to
preform multitasking and multiplexing.
The common scheduling models are:
1. Cooperative Scheduling of ready tasks in a circular queue
2. Cooperative Scheduling with Precedence Constraints
3. Cyclic scheduling of periodic tasks and Round Robin Time Slicing Scheduling
4. Pre-emptive Scheduling
5. Scheduling using ‘Earliest Deadline First’ (EDF) precedence
6. Rate Monotonic Scheduling using ‘higher rate of events occurrence First’ precedence
7. Fixed Times Scheduling
8. Scheduling of Periodic, sporadic and aperiodic Tasks
9. Advanced scheduling algorithms using the probabilistic Timed Petri nets (Stochastic) or Multi
Thread Graph for the multiprocessors and complex distributed systems

3. CYCLIC SCHEDULING
 A cyclic scheduling model is a scheduling concept in which a set of activities are
repeated an indefinite number of times.
 Let us assume that we have periodically occurring three tasks.
 Let the first task execute at time frames t1,t1+(Tcycle), t1+(2Tcycle) …
The second task occurs at t2, t2 +(Tcycle), t2+(2Tcycle) …
The third task occurs at t3, t3+Tcycle, t3+(2Tcycle)…
 Tcycle is the time for repeating the cycle of execution of tasks in order 1,2,3 and
equals start of task 1 time frame to end of task 3 time frame.
 Tcycle is period after which each task time frame allotted to that repeats.
 Each task has the same priority for execution in the cyclic mode.
 Cycling scheduling is very efficient for handling periodic tasks when the number of
tasks is small.

4.  An example of cyclic scheduling is Orchestra Playing
Robots.
 First the director robot sends the musical notes. Then,
the playing robots receive and acknowledge to the
director.
 In next cycle, the master robot again sends the musical
notes.
 The cyclic scheduler is used in this case to send and
receive signals by each robot after the cycle period.

5. ROUND ROBIN SCHEDULING
 Round robin is one of the oldest, simplest and fairest algorithm.
 Widely used model in traditional OS.
 The name of the algorithm comes from the round-robin principle known
from other fields, where each person takes an equal share of something in
turn.
 Round robin means that each ready task runs turn by turn only in a cyclic
queue for a limited time slice, Tslice.
Tslice = Tcycle/ N
N= Number of tasks
 Time slice is minimum time assigned for each task to be processed. It
can vary for different processes.
 Round robin is a hybrid model of clock-driven model (for example, cyclic
model) as well as event driven (for example, pre-emptive).
 A real time system responds to the event within a bound time limit and
within an explicit time.

6. UNDERSTANDING ROUND ROBIN WITH AN EXAMPLE
Suppose after every 20ms there is a stream of coded messages reaching Port A of
an embedded system. The multiple processes consist of 5 tasks as follows

7. Time slicing is done as follows:

8. WORST CASE LATENCY
Worst case latency is the maximum time taken for execution of all the tasks.
Tworst = {(dti+ sti + eti )1 + (dti+ sti + eti )2 +...+ (dti+ sti + eti )N-1 + (dti+ sti + eti )N} +
tISR
Where,
dt = detection time when a task is brought into the list
st = switching time from one task to another
et = task execution time
tISR = sum of all execution times for the ISRs

9. CASE 1
 Let us consider the case where each task is executed once and finishes off in
one cycle itself.
 When a task finishes the execution before the maximum time it can take,
there is a waiting period between two cycles.
 The worst-case latency for any task is then, Tworst = N X tslice .
 A task may periodically need execution. The period for the required repeat
execution of a task is an integral multiple of tslice.

10. CASE 2: ALTERNATIVE MODEL STRATEGY
 In Case 2 we have certain tasks which are executed more than once and do
not finish in one cycle.
 Alternate model strategy includes decomposition of a task that takes the
abnormally long time to be executed.
 The decomposition can be into two or four or more tasks.
 Then one set of tasks (say the odd numbered tasks) can run in one time slice,
t'slice and the another set of tasks (say the even numbered tasks) can run in
another time slice, t''slice .

11. CASE 3: DECOMPOSITION INTO A NUMBER OF
SEQUENTIAL STATES
 Another alternate strategy is decomposition of the long time taking task into
a number of sequential states or a number of node-places and transitions as
in finite state machine (FSM).
 Then one of its states or transitions can run in the first cycle, the next state
in the second cycle and so on.
 This task then reduces the response times of the remaining tasks that are
executed after a state change.

12. VoIP EXAMPLE
 Consider a VoIP (Voice Over Internet Protocol)
router. It routes the packets to N destinations from
N sources. It has N calls to route.
 Each of N tasks is allotted a time slice and is
cyclically executed for routing packet from a
source to its destination.
 This is an example of round robin with time slicing
scheduling.

13. ADVANTAGES OF ROUND ROBIN SCHEDULING
 It is a simple scheduling algorithm and easy to implement in software.
 Round robin scheduling is fair in the sense that every process gets an equal
share of the CPU.
 If we know the number of processes on the run queue, we can know the
worst-case response time for a process.
 Every process will be executed by CPU for fixed interval of time (which is set
as Tslice ). So in this way no process left waiting for its turn to be executed by
the CPU.

14. DISADVANTAGES OF ROUND ROBIN SCHEDULING
 Sometimes 2 processes may have the same priority, but one of them might get
suspended in the middle of execution, as it works on first come first serve
basis.
 If slicing time is too low, the processor output is reduced. More time is spent
on context switching.

15. QUESTIONS?

16. SEMINAR BY:
SIRIGAURI. N
RAGHAV. S