CHAPTER FIVE

REAL-TIME OPERATING SYSTEMS

RTOS
What’s Real time ?
 Real time in operating systems:
 The ability of the operating system to provide a
required level of service in a bounded response
time.
RTOS

 It responds to the inputs immediately(Real-

Time).
 Here the task is completed within a specified
time delay.
 In real life situations like controlling traffic signal
or a nuclear reactor or an aircraft, the operating
system has to respond quickly.
CHARACTERISTICS OF RTOS
Types of operating systems

 General purpose OS(GPOS)

Optimized for throughput and fairness
 Real-Time OS (RTOS)
Optimized for handling tasks within timing
limits
Types of RTOS
Hard RTOS
 has time-critical deadlines that must be met at the
time of request
 Strict timing requirements
 Optimized w.r.t. predictability and timing limits
 Does not share critical resources such as CPU
 Missing a single deadline cause a catastrophic
system failure
HARD RTOS
 Examples of application under hard real time
 Defense applications
 Life supporting medical equipment
 Industrial control systems
 air traffic control
 vehicle subsystems control
 Nuclear power plant control, etc…
SOFT RTOS
 Relative tolerable timing deadlines
 Tolerance is defined as a part of policy
 Occasional deadline missing may not cause
system failure
Examples
 Audio processing
 Real-time video processing
 Network interface subsystem
 Task/Process States

• Each task/Process can belong to one and only one

state at a time
• The Scheduler only operates on the processes in the
Ready state
• There is a single process in the Run/current state at
any time.
• Transitions to and from the Ready queue are affected
as a part of the execution of the RTOS
resource/object services or as a result of timing
events
Typical Task Operations

 creating and deleting tasks,

 controlling task scheduling, and
 obtaining task information.
 a job/task is characterized by :
 execution time
 ready time
 deadline and
 resource requirements.
TYPES OF TASK CONSTRAINTS

 Typical constraints that can be specified on real time tasks are

of three classes:

 Timing Constraints
 Precedence Relations
 Mutual Exclusion Constraints on Shared Resources.
 TIMING CONSTRAINTS

 In general, a real time task can be characterized by the following

parameters:

 Arrival Time ai: is the time at which a task becomes ready for
execution.
 Computation Time Ci: is the time necessary for the processor to
execute the task without interruption.
 Deadline di: is the time before which a task should be completed
to avoid damage to the system.
 Start Time si: is the time at which a task starts execution.
 Finishing Time fi: is the time at which a task finishes execution.
 Cont…
 Criticalness: is a parameter related to the consequences of missing
the deadline.
 Value vi: represents the relative importance of the task with respect
to other tasks in the system.
 Lateness Li: Li = fi – di represents the delay of a task completion
with respect to its deadline, it is negative if a task completes before
its deadline.
 Tardiness or Exceeding Time Ei: Ei = max(o, Li) is the time a task
stays active after its deadline.
 Laxity or Slack Time Xi: Xi = di – ai – Ci is the maximum time a
task can be delayed on its activation to complete within its deadline.
 Cont…
 Another timing characteristic that can be specified on a real time
task concerns the regularity of activation. In particular, tasks can
be defined as periodic or a-periodic.
 Periodic tasks consist of an infinite sequence of identical
activities called instances or jobs, that are regularly activated at a
constant rate.
 A-periodic tasks also consist of infinite sequence of identical
activities, however, their activations are not regular.
scheduling
 Giving an execution order for the jobs/tasks to
get the service such that various constraints are
satisfied.
 Done according to the task/job information
Scheduling policy
 Things to be considered in scheduling a real-time
system:
 Meeting the timing constraints of the system
 Preventing simultaneous access to shared resources
and devices
 Reducing the cost of context switches caused by
preemption
 Attaining a high degree of utilization while satisfying
the timing constraints of the system
 Reducing the communication cost in real-time
Scheduling in RTOS
 More information about the tasks are known
 No of tasks
 Resource Requirements
 Release Time
 Execution time
 Deadlines
 Being a more deterministic system better
scheduling algorithms can be devised.
CLASSIFICATION OF SCHEDULING ALGORITHMS

 Scheduling algorithms can be classified as follows:

 Preemptive: with preemptive algorithms, the running task can

be interrupted at any time to assign the processor to another
active task, according to a predefined scheduling policy.

 Non – Preemptive: with non – preemptive algorithms, a task

once started, is executed by the processor until completion all
scheduling decisions are taken as a task terminates its execution.
 Static: scheduling decisions are based on fixed parameters
assigned to tasks before their activation in these algorithms.
 Dynamic: scheduling decisions are based on dynamic parameters
that may change during system evolution.
 Cont…
 Clock Driven Scheduling
 Weighted Round Robin Scheduling
 (Greedy / List / Event Driven)
Clock Driven

 All parameters about jobs (release time/

execution time/deadline) known in advance.
 Schedule can be computed offline or at some
regular time instances.
 Minimal runtime overhead.
 Not suitable for many applications.
Weighted Round Robin

 Round Robin – give each task an equal share of the

processor(FIFO )
 Implemented when all tasks or threads have the same
priority level
 May be implemented within a priority range
Example use : High speed switching network QOS
guarantee.
 Not suitable for precedence constrained jobs.
Eg. Job A can run only after Job B. No point in giving
time quantum to Job B before Job A.
Priority Scheduling

 Pre-emptive priority scheduling – higher

priority tasks may interrupt lower priority tasks.
 Processor allocated to processes according to
priorities Categorized as:
Static – priorities are set before the system
begins execution(at design time)
Dynamic – priorities can be redefined at run
time(at runtime)
 Priority Scheduling
Earliest deadline first
 The Earliest Deadline First is a dynamic scheduling rule that
selects tasks according to their absolute deadlines.
 Specifically, tasks with earlier deadlines will be executed at higher
priorities.
 EDF is a dynamic priority assignment and is intrinsically
preemptive, that is, the currently executing task is preempted
whenever another periodic instance with an earlier deadline
becomes active.
 Cont…
Earliest Deadline First (EDF)
 Process with earliest deadline given highest priority
 The runnable processes are executed in the order determined
by the absolute deadlines of the processes
 the absolute deadlines are computed at run time and hence the
scheme is described as dynamic
Least Slack Time First (LSF)
 slack = relative deadline – execution left
Rate Monotonic Scheduling (RMS)
 For periodic tasks
 Tasks priority inversely proportional to it’s period
 Cont…
 This is a simple rule that assigns priorities to tasks according
to their request rates.
 Specifically, tasks with higher request rates will have higher
priorities.
 Since the periods are constant, RM is a fixed priority
assignment: priorities are assigned to tasks before execution
and do not change with time.
 Moreover, RM is intrinsically preemptive: the currently
executing task is preempted by a newly arrived task with
shorter period.
Preemption process
Resource Allocation in RTOS

 Resource Allocation
 The issues with scheduling applicable here.
 Resources can be allocated using
• Weighted Round Robin
• Priority Based
• Some resources are non pre-emptible
Example : semaphores
• Priority Inversion if priority scheduling is used
Other RTOS issues
• Interrupt Latency should be very small
– Kernel has to respond to real time events
– Interrupts should be disabled for minimum
possible time
• For embedded applications Kernel Size should be
small
– Should fit in ROM
• Sophisticated features can be removed
– No Virtual Memory
– No Protection
INTERRUPT HANDLING OF RTOS.

• An interrupt is a signal from a device attached to a

computer or from a program with in a computer
that causes the main program, that is operating
system to stop and figure out what to do next.
• Interrupts cause the processor to suspend the
operations whatever it is doing instead execute
the code that will respond to the event whatever
caused the interrupt.
APPLICATIONS OF RTOS.
 modern telecommunication systems
 Radar systems
 network switching control systems,
 satellite monitoring systems, satellite launch-
control and global positioning systems all have
their roots in RTOS.
 Now a days RTOSes have increasingly finding use
in strategic and military operations. These are
used in guided missile launching units, track-and-
trace spy satellites, etc.
RTOS vs GPOS
 Unfair scheduling  Fair scheduling
Scheduling based on  Scheduling can be
priority adjusted dynamically for
 Predictable behavior optimized throughput
 Priority inversion is a  No predictability
major issue guarantees
 Kernel is preemptive  Priority inversion remain
either completely or up unnoticed
to maximum degree  Kernel is long/non-
preemptive code sections
Limitations for RT applications
 Page faults
 Occurs when requested data is not available or its
reference is absent from a page table
 Types of page faults
 Major/Hard page faults
 Minor/Soft page faults
Major page faults
 Occurs when requested data has to be fetched
from disk
 Can cause very large latencies
 RT applications should be written to minimize
Major page faults
Minor page faults
 Occurs when requested data resides in the main
memory but missing from TLB
 It does not involve IO operation to fetch data
from disk
 Has negligible impact in RT performance
 Avoid page faults
 Use mlockall() to load all the address space of the
program and then lock it such that it cannot be
swapped out
 Create all threads at startup time of the application
because run time thread creation can cause latencies
 Avoid dynamic allocation and freeing of memory
 Minimize the use of system calls that are known to
generate page faults
Context switching
 Switching the CPU to another task requires
saving the state of the old task and loading the
saved state for the new task.
 The time wasted to switch from one task to
another without any disturbance is called context
switch or scheduling jitter.
Is context switching really a problem?