Click to edit Master subtitle style Chapter 19 Real-Time Systems CS 540 Advanced Operating systems Instructor: Dr. Behzad Perviz Fall 2010 Presented By:

Click to edit Master subtitle style

Chapter 19Real-Time Systems

CS 540 Advanced Operating systems Instructor: Dr. Behzad Perviz

Fall 2010

Presented By: Monali Bhavsar Amee Joshi

Real-Time Systems

System CharacteristicsFeatures of Real-Time SystemsImplementing Real-Time

Operating SystemsReal-Time CPU SchedulingAn Example: VxWorks 5.x

ObjectivesTo explain the timing requirements of

real-time systemsTo distinguish between hard and soft

real-time systemsTo discuss the defining characteristics

of real-time systemsTo describe scheduling algorithms for

hard real-time systems

Overview of Real-Time SystemsA Real-time system requires that results be

produced within a specified deadline period ◦ Example: Robot◦ Contrast: Desktop computer system, Batch

processing system.

An Embedded system is a computing device that is part of a larger system.(i.e. automobile, airliner)

◦ No timing requirements.◦ Embedded in specialized devices.◦ Presence of computing device is not

obvious.◦ Examples: dishwashers, microwave ovens,

cameras, MP3 players.

Continue…A Safety-critical system is a real-

time system with catastrophic results in case of failure.◦ Weapons systems, antilock brake system, flight management system & health related systems.

◦ System must respond to events by specific deadline period.

A Hard real-time system guarantees that real-time tasks be completed within their required deadlines.◦ Critical real time tasks be completed within their deadlines.

◦ Safety critical systems are typically hard real time systems.

A Soft real-time system provides priority of real-time tasks over non real-time tasks.◦ Priority retain until task completed.

◦ Linux and many OS provide soft real time system.

System Characteristics

Single purposeSmall sizeInexpensively mass-producedSpecific timing requirements

System CharacteristicsSingle purpose

◦ Unlike PC’s real time systems serves single purpose. Design of it reflects single purpose and its simple.

Small size◦ Existing environment is constrained

in physical space so CPU power and memory available is less then standard pc’s.

◦ Real time systems run on 8- or 16- bit processors and less then megabytes of memory.

◦ Footprint: amount of memory required to run the OS and its applications. Real time systems must have small

footprints.

Specific timing requirements◦ Real time operating systems meet

timing requirements by using scheduling algorithms that gives real-time processes the highest scheduling priorities.

◦ Priority of scheduling tasks does not degrade over time.

◦ Technique for addressing timing requirements: minimize response time to events such as Interrupts.

Click to edit the outline text format Second

Outline Level Third

Outline Level

Fourth Outline Level

Fifth Outline Level

Sixth Outline Level

Seventh Outline Level

Eighth Outline Level

Ninth Outline LevelClick to edit Master text styles

◦ Second level Third level

Fourth level Fifth level

System Characteristics Inexpensively mass-produced

◦ Real time systems are used in home appliances and consumer devices which are cost conscious environment, so microprocessors for real time systems must inexpensively mass produced.

◦ Example: SOC.

Bus Oriented System

System-on-a-Chip

Many real-time systems are designed using system-on-a-chip (SOC)strategy

SOC allows the CPU, memory, memory-management unit, and attached peripheral ports (I.e. USB) to be contained in a single integrated circuit.

Less expensive then bus oriented organization.

System-on-a-Chip

Features of Real-Time KernelsFeatures provided by many

operating systems are:◦ Support variety of peripheral devices◦ Protection and security mechanism◦ Multiple users

Supporting these features results in large kernel. Example, Windows XP.

Most real-time systems do not provide the features found in a standard desktop system as above.

Reasons include◦ Real-time systems are typically single-

purpose◦ Real-time systems often do not

require interfacing with a user◦ Features found in a desktop PC

require more substantial hardware that what is typically unavailable in a real-time system due to lack of memory and fast processors. Both of these are unavailable in real time

systems due to space constraints. Addition to that many systems lack

sufficient space to provide graphical displays or disk drives, they support file systems using NVRAM(Non Volatile RAM).

◦ Features of desktop PC increase the cost of real time systems which makes systems economically impractical.

Virtual Memory in Real-Time Systems Providing virtual memory features requires that

the system include a Memory Management Unit(MMU).◦ MMUs increase the cost and power consumption.◦ Time required to translate logical address to

physical address especially in case of Translation Look aside Buffer(TLB) miss – may be prohibited in hard real time systems.

Address translation may occur via:1) Real-addressing mode where programs generate

actual addresses

2) Relocation register mode

3) Implementing full virtual memory

Address Translation


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Address TranslationReal-addressing mode:

◦ CPU generates logical address L which must be

mapped to physical address P.

◦ Bypass the logical address and directly generate physical address.

◦ Not employ virtual memory techniques so P equals L.

◦ Problem: no memory protection between processes and programmers need to specify physical location of memory load.

◦ Benefits: fast, no time spent on address translation.

◦ Used in embedded systems with hard real time constraints.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Address TranslationRelocation register mode:◦ Same as Dynamic relocation register.

◦ Relocation register R is set to memory location where a program is loaded.

◦ Physical address P is generated by adding the contents of relocation register R to L.

◦ Real time systems are configure the MMU to perform this way because MMU can easily translate logical addresses to physical addresses using P=L+R.

◦ This system will also not provide memory protection between processes.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Address Translation Implementing full virtual memory:

◦ Address translation take place via page table and

Translation Look aside buffer(TLB).

◦ This strategy provides program to be loaded at any memory location & memory protection between two processes.

◦ Without attaching disk drives not possible to provide all virtual memory features like Demand Paging and Swapping. Contrast to that some system provides

that using NVRAM. Examples: LynxOS and OnCore

Systems.

Implementing Real-Time SystemsIn general, real-time operating

systems must provide:1)Preemptive, priority-based

scheduling2)Preemptive kernels3)Latency must be minimized

Implementing Real-Time Systems Preemptive, priority-based

scheduling:◦ Important feature: must respond

immediately to a real time process so system must support Priority-based algorithm with Preemption.

Priority based scheduling algorithms assign priority based on their importance.

If scheduler supports preemption, a process currently running on CPU will be preempted if a higher priority process will available to run.

◦ Solaris, Windows XP & Linux systems assign highest scheduling priority to real time processes.

◦ This will provide only soft real time functionality. For hard real time functionalities we need additional scheduling features to meet timing requirements.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Windows XP priorities:Windows XP has 32 different priority levels, the

highest priority level values 16-31 are reserved for real time processes.

Implementing Real-Time Systems Preemptive kernels:

◦ Allows the Preemption of a task running in kernel mode.

◦ Designing preemptive kernel is difficult so if quick response is not require its not implemented. Ex. Windows XP is non-Preemptive.

◦ In Hard real time systems preemptive kernels are mandatory.

◦ There are two strategies to make kernel preemptible:

First: Insert Preemption points in long duration system calls.

Preemption points can be placed at safe locations in kernel that is where kernel data structure is not modified.

Second: Use of synchronization mechanisms.

Any kernel data being updated are protected from modification by the high priority process so kernel will always preemptible.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Minimizing LatencyEvent latency is the amount of time from

when an event occurs to when it is serviced.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Interrupt Latency Interrupt latency is the period of time

from when an interrupt arrives at the CPU to when it is serviced.

Interrupt Latency Important factor contributing to interrupt latency

is the amount of time interrupts may be disabled while kernel data structures are being updated.

Real time operating systems required that interrupts be disabled for very short period of time.

In hard real time systems it must not only be minimized, it must in fact bounded to guarantee the deterministic behavior of hard real time systems.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Dispatch LatencyDispatch latency is the amount of time

required for the scheduler to stop one Process and start another

Dispatch LatencyIf we want to provide real time

tasks with immediate access to CPU mandates that operating system should minimize Dispatch latency. To keep Dispatch latency low, provide Preemptive kernels.

The Conflict phase of dispatch latency has two components:◦ Preemption of any process running

in the kernel◦ Release by low priority processes of

resources needed by a high-priority process. Ex. Solaris.

One issue that can affect the dispatch latency arises when a higher priority process needs to read or modify kernel data that are currently being accessed by a lower priority process. Or a chain of lower priority processes.

Problem: Priority inversion.Solved by: Priority-inheritance

protocol.

Priority Inversion


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Priority Inversion & Priority Inheritance

Real-Time CPU Scheduling Scheduling : Deciding how to allocate a single

resource

among multiple clients.◦ In what order and for how long.◦ Usually refers to CPU scheduling.

CPU Scheduling decisions may take place when a process:◦ Switches from running to waiting state◦ Switches from running to ready state◦ Switches from waiting to ready◦ Terminates.

Real-Time CPU SchedulingTwo types of Real Time

◦ Soft Real Time : Meet Deadline most of the time, but not mandatory.

Example : Live audio-video systems are usually soft real-time; violation of constraints results in degraded quality, but the system can continue to operate.

◦ Hard Real Time : Must meet deadline, otherwise can cause fatal error.

Example : a car engine control system is a hard real-time system because a delayed signal may cause engine failure or damage.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Real-Time CPU SchedulingPeriodic processes require the CPU at specified intervals

(periods)p is the duration of the periodd is the deadline by when the process must be

servicedt is the processing time

Real-Time CPU Scheduling Unusual about this scheduling is that a process

may have to announce its deadline requirements to the scheduler.

Using Technique an Admission Control algorithm the scheduler◦ either admits the process, guaranteeing that the

process will complete on time,◦ or rejects the request as impossible if it cannot

guarantee that task will be serviced by its deadline.

Real Time CPU Scheduling Algorithms

Rate Monotonic SchedulingEarliest Deadline First SchedulingProportional Share SchedulingPthread Scheduling

Rate Monotonic Scheduling It schedules periodic tasks using a static

priority policy with preemption.◦ If a lower priority process is running and a

higher priority process becomes available to run, it will preempt the lower priority process.

Each periodic task is assigned a priority inversely based on its period:◦ The shorter the period , the higher the priority.◦ The longer the period, the lower the priority.

Rate monotonic scheduling assumes that the processing time of a periodic process is the same for each CPU burst.◦ Every time a process acquires the CPU, the

duration of its CPU burst is the same.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Example Two Processes P1 and P2. The periods for p1 = 50 and p2 = 100. The Processing times are t1 = 20 for P1 and t2 = 35 for

P2. The deadline for each process requires that it complete

its CPU burst by the start of its next period. The CPU utilization of a process Pi as the ratio of its burst

to its period – ti/pi so, for P1 it is 20/50 = 0.40 and for P2 it is 35/100 = 0.35. so total CPU utilization of 75 percent.

First, suppose we assign P2 a higher priority than P1.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Example - continue

Now suppose we use rate monotonic scheduling, in which we assign P1 a higher priority than P2.


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Missing Deadline with rate monotonic scheduling Assume that Process P1 has a period of p1 = 50

and CPU burst of t1 = 25. For P2, the corresponding values are p2 = 80

and t2 = 35. The total CPU utilization of the two processes is

(25/50) + (35/80) = 0.94.

Limitation CPU utilization is bounded.

The worst case CPU utilization for scheduling N processes is 2(2^(1/n)-1).

◦ With one process in the system, CPU utilization is 100 percent, but it falls to approximately 69 percent as the number of processes approaches infinity.

◦ With two processes, CPU utilization is bounded at about 83 percent.

Earliest Deadline First Scheduling

It dynamically assigns priorities according to deadline.◦ The earlier the deadline, the higher the priority.◦ The later the deadline, the lower the priority.◦ If two tasks have the same absolute deadlines,

chose one of the two at random (ties can be broken arbitrarily).


Outline Level Third

Outline Level


Fifth Outline Level

Sixth Outline Level






Example

Suppose we have two Process P1 and P2. P1 has values of p1 = 50 and t1 = 25. P2 has values of p2 = 80 and t2 = 35.

Unlike the rate monotonic algorithm, EDF scheduling does not require that processes be periodic, nor must a process require a constant amount of CPU time per burst..◦ The only requirement is that a process announce its

deadline to the scheduler when it becomes runnable.

The appeal of EDF scheduling is that, theoretically it can schedule processes so that each process can meet its deadline requirements and CPU utilization will be 100 percent.◦ It’s impossible due to the cost of context switching

between processes and interrupt handling.

Proportional Share Scheduling

Associate a weight with each application and allocate CPU bandwidth proportional to weight

T shares are allocated among all processes in the system

An application receives N shares where N < T

This ensures each application will receive N / T of the total processor time

Wt=2

2/3 1/3

Wt=1

Applications

CPU bandwidth

Example

Assume that a total of T = 100 shares is to be divided among three processes, A, B and C.

A is assigned 50 shares, B is assigned 15 shares, and C is assigned 20 shares.

This scheme ensures that A will have 50 percent of total processor time, B will have 15 percent, and C will have 20 percent.

Proportional Share Scheduling

Proportional Share Schedulers must work in conjunction with an admission control policy to guarantee that an application receives its allocated shares of time.◦ An admission control policy will only admit a client

requesting a particular number of shares if there are sufficient shares available.

Pthread Scheduling The Pthread (POSIX Thread) Library is set of

functions that enable C/C++ code to spawn multiple “threads” of execution to do multiple tasks simultaneously.

The Pthread API provides functions for managing real-time threads.

Pthreads defines two scheduling classes for real-time threads:

(1) SCHED_FIFO - threads are scheduled using a FCFS strategy with a FIFO queue. There is no time-slicing for threads of equal priority

(2) SCHED_RR - similar to SCHED_FIFO except time-slicing occurs for threads of equal priority

Pthread Scheduling API#include <pthread.h>#include <stdio.h>#define NUM THREADS 5int main(int argc, char *argv[]){

int i, policy;pthread_t tid[NUM THREADS];pthread_attr_t attr;

/* get the default attributes */pthread _attr_ init(&attr);

/* get the current scheduling policy */if (pthread_attr_getschedulepolicy (&attr,

&policy) != 0)fprintf (stderr, “Unable to get policy.\n”);

Pthread Scheduling APIelse {

if (policy == SCHED_OTHER)printf(“SCHED_OTHER\n”);

elase if (policy == SCHED_RR)printf(“SCHED_RR\n”);

else if (policy == SCHED_FIFO)printf(“SCHED_FIFO\n”);

}

/* set the scheduling policy - FIFO, RT, or OTHER */

if (pthread_attr_setschedpolicy (&attr, SCHED_OTHER) != 0)

/* create the threads */for (i = 0; i < NUM _THREADS; i++)

pthread _create (&tid[i], &attr, runner, NULL);

Pthread Scheduling API/* now join on each thread */

for (i = 0; i < NUM _THREADS; i++)

pthread _join (tid[i], NULL);

}

/* Each thread will begin control in this function */

void *runner(void *param)

{

/* do some work … */

pthread _exit(0);

}

An Example: VxWorks 5.x A popular real-time operating system providing

hard real-time support. Commercially developed by Wind River Systems. It is widely used in automobiles, consumer and

industrial devices, and networking equipment such as switches and routers.

It is also used to control the two rovers – Spirit and Opportunity – that began exploring the planet Mars in 2004.

The Organization of VxWorks

Wind Microkernel The Wind microkernel provides support for the

following:

(1) Processes and threads using Pthread API;

(2) preemptive and non-preemptive round-robin scheduling;

(3) manages interrupts (with bounded interrupt and dispatch latency times);

(4) shared memory and message passing as communication between separate tasks. Also allows tasks to communicate using a technique known as pipes. Also provides semaphores and mutex locks with a priority inheritance protocol to prevent priority inversion.

Interesting approach to memory management Supports two levels of virtual memory.

First Level: ◦ Quite simple, allows for control of the cache on a

per-page basis.◦ Enables and application to specify certain pages as

non-cacheable.

Second Level: ◦ Virtual memory requires the optional virtual

memory component VxVMI along with processor support for a memory management unit(MMU).

◦ VxWorks allows pages containing kernel code along with the interrupt vector to be declared as read-only.

References

Operating System Concepts, 8th edition by Silberschatz, Galvin and Gange.

www.wikipedia.com

Thank you

Documents

Click to edit Master subtitle style Chapter 19 Real-Time Systems CS 540 Advanced Operating systems Instructor: Dr. Behzad Perviz Fall 2010 Presented By: