Operating System: Management and Functions

The Operating System

• A computer system needs hardware and software resources to complete tasks
• The required resources include input/output devices, memory, file storage, CPU etc.
• Operating systems manage these resources and allocate them to specific programs and users when necessary
• Operating systems manage processor, memory, files, and I/O devices internally
• An operating system is the interface between the user and the machine

Operating System Management Tasks

• Manages the processor by ordering tasks and pairing them into manageable sizes for the CPU
• Manages memory by coordinating data flow to and from RAM, determining the need for virtual memory
• Manages devices by providing an interface between connected devices
• Manages storage by directing permanent data storage
• Applications enable standard communication between software and computers

Functions of Operating System

• Boots the computer
• Performs basic computer tasks, such as managing peripheral devices like the mouse and keyboard
• Provides a user interface, be it command line or GUI
• Handles system resources like memory and CPU time allocation
• Provides file management to manipulate, store, retrieve, and save data
• Handles errors by taking preventive measures

History of Operating Systems

• The evolution is dependent on the development of computer systems and their usage

Operating Systems - Late 1950s

• Operating systems were improved and started supporting certain usages by the late 1950s:
• Single-stream batch processing was possible
• Common, standardized input/output routines could be used for device access
• Program transition capabilities to reduce the overhead of starting a new job were implemented
• Error recovery to clean up after abnormally terminated jobs was added
• Job control languages allowed users to specify job definition and resource requirements

Operating Systems - In 1960s

• 1961: The dawn of minicomputers
• 1960s: Disks became mainstream
• 1966: Minicomputers became cheaper, more powerful, and more useful
• 1967-1968: The mouse was invented

Supported OS Features by 1970s

• Multi-user and multitasking capabilities were introduced
• Dynamic address translation hardware and virtual machines emerged
• Modular architectures came into existence
• Personal, interactive systems came into existence

Accomplishments after 1970

• 1971: Intel announces the microprocessor
• 1972: IBM introduces VM: the Virtual Machine Operating System
• 1973: Ethernet
• 1974: The Personal Computer Age begins; Gates and Allen write BASIC for the Altair
• 1976: Apple II
• August 12, 1981: IBM introduces the IBM PC
• 1983: Microsoft begins work on MS-Windows
• 1984: Apple Macintosh comes out
• 1990: Microsoft Windows 3.0 comes out
• 1991: GNU/Linux
• 1992: The first Windows virus emerges
• 1993: Windows NT
• 2007: iOS
• 2008: Android OS

Types of Operating Systems

• Simple Batch System
• Multiprogramming Batch System
• Multiprocessor System
• Desktop System
• Distributed Operating System
• Clustered System
• Realtime Operating System
• Handheld System

Simple Batch Systems

• There is no direct interaction between the user and the computer in a simple batch system:
• The user submits a job using cards or tape to a computer operator
• The computer operator assembles a batch of jobs on an input device
• Jobs are batched based on programming language and other requirements
• A special program, the monitor, oversees each program's execution
• The monitor resides in the main memory and is always available for execution

Advantages of Simple Batch Systems

• No interaction between user and computer
• No mechanism to prioritize processes

Multi-User Operating Systems

• Allow many users to take advantage of the computer's resources simultaneously
• The OS ensures balanced requirements for the users and separate resources for programs to prevent one user's problem from affecting others
• Unix, VMS, and mainframe OS like MVS are examples of multi-user operating systems
• Windows 2000 and Novell Netware can support many networked users but they aren't considered true multi-user operating systems

Functions of the Operating System

• A bootstrap is the program that initializes the operating system (OS) during startup
• Bootstrapping loads a set of instructions when a computer is turned on or booted
• During startup, diagnostic tests like POST are performed to check configurations for devices
• POST also implements routine testing for the connection of peripherals, hardware, and external memory devices
• The bootloader or bootstrap program is then loaded to initialize the OS

Process Management

• A process is a program in execution but is an 'active' entity as opposed to the program, which is considered a 'passive' entity
• Attributes held by processes include hardware state, memory, CPU etc.

Process States

• NEW: The process is being created
• READY: The process is waiting to be assigned to a processor
• RUNNING: Instructions are being executed
• BLOCKED: The process is waiting for an event, like I/O completion or a signal
• TERMINATED: The process has finished execution

Process Control Block

• A Process Control Block exists for each process, containing all information about the process
• It is a data structure containing:
• Process State: Its current condition (running, waiting, etc.)
• Process ID and the parent process ID
• CPU registers and Program Counter: CPU registers and program counter hold the address of the next instruction to be executed
• CPU Scheduling information: Priority and pointers to scheduling queues
• Memory Management information: Page tables or segment tables
• Accounting information: User and kernel CPU time consumed, account numbers, limits, etc.
• I/O Status information: Devices allocated, open file tables, etc.

What is a Deadlock?

• Deadlocks occur when a set of processes are blocked, each holding a resource and waiting to acquire a resource held by another process.

How to avoid deadlocks

• Deadlocks can be avoided by evading at least one of the four conditions:
• Mutual Exclusion
• Hold and Wait
• No Preemption
• Circular Wait

Handling deadlock

• Focuses on preventing deadlocks
• The strategies that can be used to remove deadlock after occurrence include:
• Preemption
• Rollback
• Kill one or more processes

Process Scheduling & Scheduling Algorithms

• Determining which process in the "ready" state should move to the "running" state is known as Process Scheduling
• Keeping the CPU busy and minimizing response time for programs is the primary goal
• Achieved by applying rules for "swapping" processes in and out of the CPU
• Scheduling falls into non-pre-emptive scheduling and pre-emptive scheduling
• Non Pre-emptive Scheduling: current process release CPU voluntarily
• Pre-emptive Scheduling: OS allocates CPU

Pre-emptive Scheduling

• Shortest-Job-First (SJF),
• round robin

Non-pre-emptive

• First come first serve (FCFS)
• Shortest-Job- First (SJF)

Scheduling Metrics

• ThroughPut: the number of processes completed per unit of time
• Turnaround Time: The interval from submission to completion
• Waiting Time: the period spent waiting in the ready queue
• Load Average: the average number of the processes residing in the ready queue waiting for their turn to get into the CPU
• Response Time: the amount of time from request submission until the first response is produced

Aim of Scheduling Algorithms

• Maximize CPU utilization
• Ensure fair allocation of CPU
• Maximize number of processes that complete their execution per time unit
• Minimize time taken by a process to finish execution
• Minimize time a process spends waiting in the ready queue
• Minimize time it takes for a process to produce its first response
• Ensure no task is starved of resources

Shortest Job First (SJF) Scheduling

• Schedules processes based on burst time or duration, executing shortest first
• This approach minimizes waiting time, typically for a batch system
• Successfully executed if the burst time or execution duration of the processes is known in advance
• It of 2 types:
  • Non-pre-emptive
  • Pre-emptive

Problem with Non-pre-emptive SJF

• Can lead to the problem of starvation, where shorter processes have to wait unnecessarily

Pre-emptive Shortest Job First

• Jobs are put into a ready queue when they arrive
• When a process with a shorter burst time arrives, the existing one is preempted or removed from execution
• The job with a shorter burst time will then execute first

Round Robin Scheduling

• Uses a single queue of processes
• When the system timer fires, the next process in the queue is switched to, and the preempted process returns to the queue.
• A fixed amount of time is allotted to each process, and is called a quantum
• After a process has executed for a given time period, it is then preempted and another process executes
• Context switching saves states of preemptied processes

First Come First Serve (FCFS) Scheduling

• The process that requests the CPU first is allocated the CPU first
• This resembles a FIFO queue data structure
• Easy to understand and implement programmatically
• Real life example of this is buying tickets at ticket counter

Problems with FCFS Scheduling

• It is a non-preemptive algorithm which means the process priority doesn't matter
• Resources utilization in parallel is not possible, which leads to Convoy Effect, and hence poor resource(CPU, I/O etc) utilization

Priority Scheduling

• A priority is assigned for each process, and processes with the highest priority are executed first
• Processes with the same priority are executed in FCFS manner
• Decisions can be made based on memory, time, or any other resource requirements

Comparison of Scheduling Algorithms

• CPUs apply different scheduling algorithms to schedule processes

First Come First Serve (FCFS) - Advantages

• The algorithm doesn't include any complex logic
• It's pretty simple and easy to implement
• Every process will eventually get a chance to run, preventing starvation

First Come First Serve (FCFS) - Disadvantages

• There is no option for pre-emption of a process
• If a process executes for a long time, other processes will have to wait longer

Shortest Job First (SJF) - Advantages

• Processes are executed starting with the shortest jobs before running the longer ones
• Higher throughput because more processes can be executed in a shorter period of time

Shortest Job First (SJF) - Disadvantages

• The time taken by a process must be known by the CPU beforehand, which is not possible
• Longer processes will have more waiting time, which can eventually cause starvation

Round Robin (RR) - Advantages

• Each process is served by the CPU for a set amount of time, so each process is given the same priority
• Starvation doesn't occur

Round Robin (RR) - Disadvantages

• The throughput relies heavily on the length of the time quantum
• If the time quantum is longer than needed, it will behave like FCFS
• If the time quantum is shorter than needed, the CPU will switch too frequently between processes, reducing CPU efficiency

Priority based Scheduling - Advantages

• The priority of a process can be selected on memory requirement, time requirement or user preference
• A higher-end game will have better graphics

Priority based Scheduling - Disadvantages

• A second scheduling algorithm is required to schedule the processes that have the same priority
• If low priority processes keep waiting for higher priority processes, starvation will occur

Usage of Scheduling Algorithms in Different Situations

• Every scheduling algorithm has a situation where it is the best choice

Situation 1

• The incoming processes are short and there is no need for the processes to execute in a specific order
• FCFS is the best in this case

Situation 2

• The processes are a mix of long and short processes and the task will only be completed if all the processes are executed successfully in a set time
• Round Robin Scheduling is the best in this case

Situation 3

• The processes are a mix of user based and kernel based processes
• Priority based scheduling is most efficient

Interrupts

• A processor has to wait for a signal to begin processing but now only does it when new data is available
• The signal from hardware or software should have the highest priority over all other instructions in the system
• This processing signal is called Interrupt

What is an Interrupt?

• It is a signal from hardware or software with the highest signal priority that a processor should process immediately

Types of Interrupts

• There are 2 types:
• Hardware Interrupts: If the signal for the processor is from an external device
  • Maskable Interrupt: Hardware interrupts which can be delayed when a much higher priority interrupt occurs to the processor
  • Non-Maskable Interrupt: Hardware interrupts that cannot be delayed
• Software Interrupts: Interrupts that software instructions caused
  • Normal Interrupts: Software interrupts caused by software instructions
  • Exception: When a software instructions causes an unplanned interrupts

Interrupt Handling

• The instruction cycle consists of fetch, decode, execute and read/write functions After every instruction, the processor will check for interrupts with the instruction register If there is an interrupt, it triggers which will stop the current instruction to save the configuration
• Its configuration will load to the program counter of the interrupt from a location given by the interrupt vector table
• The interrupt handler will load the instruction and its configuration from the saved register, and then complete This saving is called context switching
• The interrupt handler is also called as Interrupt service routine (ISR).
• Interrupts can occur at any time and are asynchronous. ISRs can call asynchronous interrupts
• The interrupt service mechanism can call the ISR's from multiple sources
• ISR's can handle both maskable and non-maskable interrupts, program can disable or enable an interrupt handler call
• The nested interrupts are allowed in ISR for diversion to other ISR

Memory Management & Device Management

• The process of controlling, coordinating, and assigning computer memory to various running programs to optimize overall system performance

Virtual Memory

• Combines active RAM and inactive memory on Secondary Storage to form a large range of contiguous addresses
• The primary benefits of virtual memory include freeing applications from having to manage a shared memory space, increased security due to memory isolation, and paging
• The OS divides memory into pagefiles or swap files with a fixed number of addresses
• Each page is stored on a disk so when the page is needed
• The portion of the hard disk that acts as physical memory is called a page file
• A page fault is a type of exception by the computer's hardware
• Page faults usually mean that they need to allocate more memory

Thrashing

• When the computer transfers pages it causes slowing down normal work
• In computer science, thrashing occurs when a computer's virtual memory resources are overused
• This inhibits the user from working

Device Management

• The process of managing the implementation, operation and maintenance of physical devices

Device Drivers

• A device driver is a computer program that operates or controls a particular type of device attached to a computer
• It provides a software interface to hardware devices
• They provide the interrupt handling required for any necessary asynchronous time-dependent hardware interface.

Devices, Interrupt Handling & the PCB

• PCB is a data structure in the operating system kernel containing the information needed to manage the scheduling of a particular process
• An interrupt is a signal sent to the processor because of a an event for immediate attention
• When a device needs attention or has data, the operating system may pause the running process
• After IRQ has been handled, OS then returns to the interrupted process to continue

Input/Output Control & Peripheral Control

• Used as an input/output interface, IOC, or PIOC for Peripheral input/output controller
• The input/output controller is a device that interfaces between an input or output device and the computer or hardware device
• The input/output controller is located on the motherboard
• The controller is a chip, card, or tool that interfaces with devices

Polling

• It is continuously checking other programs or devices by a program or device
• Polling is the opposite of interrupting

Buffer

• It a memory area that stores data while they are transferred between two devices
• Buffer usage occurs when the process has to
  • Enable faster transfer
  • Enable copy for application IO
• Adapt IO between devices

Spooling

• Sending data (intended for printing or processing on a peripheral device) to an intermediate store (e.g. buffer) is spooling
• Spooling allows programs to "hand off" work to be done by the peripheral
• Spooling is a combination of buffering and queueing

File Management

• Manages how data is sorted into files, it helps the user to see files
• The OS shows and helps store files, and fixes any errors
• The OS alerts the user of corrupted files

File

• Related documents that is on the computer
• Example of files are disks and tapes
• They may hold different records, such as images

Directories / Folders

• Files are stored on the folder, and the folders are structured like a tree
• Example file systems includes Linux, MS-DOS, OS/2, and Unix

Security of Files

• Security on the computer to manage access
• User ID: Each file is owned by a user
• Passwords: To secure from unauthorized users

Lockwords

• In security a lockword is a password associated with a file or data set, only set/changed by the owner
• Not used to encrypt files it meant to be shared

Access Control List

• Grants permissions access to another user Is a list showing which users are granted permission

File Encryption

• Providing file security, only authenticated users can access the data
• If a user tries to gain access without the permission

File Compression

• Compress an archive to save space on the hard drive

Activity Logs

• Records user actions. What a user did and when