Operating Systems: Introduction and Overview

Lecture 1: Introduction & Overview of Operating Systems

• The lecture introduces fundamental concepts of operating systems
• The lecture is presented by Dr. Nasser Tamim

Objectives

• The purpose is to describe the basic organisation of computer systems
• The goal is to provide a tour of the major components of systems
• The aim is to give an overview of computing environments
• The lecture will explore open source operating systems

What is an Operating System?

• An operating system is a program that acts as an intermediary between a computer user and the computer hardware
• The operating system enables execution of user programs (apps) and simplifies problem-solving
• Operating systems make computers convenient to use by creating interaction between user and hardware, working as a platform for applications
• The operating system handles resources efficiently

Computer System Structure

• A computer system consists of four components
• The Hardware provides computing resources such as CPU, memory, and I/O devices
• The Operating System controls and coordinates hardware use across applications and users
• Application Programs define how system resources are used to solve users' computing problems
  • Examples are word processors, compilers, browsers, database systems, and video games
• Users relates to entities, including people, machines, and other computers, that utilize the system

Operating System Roles

• User perspective is convenience, ease of use, and performance, regardless of resource utilization
  • The OS balances performance and resource utilization
• Shared computers serve many users, so the OS manages services and optimizes resource allocation to ensure quality
• Dedicated systems use dedicated resources and shared server resources
  • The OS provides services with optimal resource utilization on-demand
• Handheld computers have limited resources
  • The OS balances performance with minimal power consumption for longer battery life
• Embedded computers without user interfaces require the OS to provide required services without intervention

Operating System Definition

• An OS is a resource allocator that manages all resources and decides between conflicting requests for efficient utilization
• An OS is a control program that prevents errors and improper computer use
• There is no universally accepted OS definition
• One approximation is "Everything a vendor ships when you order an operating system," but capabilities and functionalities vary wildly across operating systems
• The Kernel is the core program running at all times on the computer
• All else is either a system or application program

Computer Startup

• A Bootstrap program is loaded during power-up or reboot with the following characteristics
• Stored in ROM or EPROM, also known as firmware
• Initialises OS, hardware, and system programs
• Loads the operating system kernel and starts execution

Computer System Organization

• One or more CPUs and device controllers connect through a common bus
• The common bus provides access to shared memory (main memory) for executing operations or tasks
• CPUs and devices compete for memory cycles concurrently, turns or time is needed to perform a task

Computer-System Operation

• I/O devices and the CPU can execute concurrently
• Each device controller manages a specific device type has a local buffer
• The local buffer stores for a certain amount of time
• The CPU moves data between main memory and local buffers
• I/O moves data between device and the controller's local buffer
• A device controller informs the CPU of operation completion via an interrupt

Common Functions of Interrupts

• Interrupts transfer control to the interrupt service routine (ISR) via an interrupt vector with addresses of all service routines
• Interrupt architecture saves the address of the interrupted instruction
• A trap or exception is a software-generated interrupt that signals errors or user requests
• Operating systems are interrupt driven

Interrupt Handling

• To handle interrupts, the OS preserves the CPU by storing registers to allow for Program Counter (PC) management
• Polling allows checking device statuses
• Vectored interrupt systems assign unique codes, four to eight bits long, to interrupting devices
  • The device sends its code over the data bus to tell the processor which routine to execute
• Separate code segments handle actions for each type of interrupt

I/O Structure

• Control returns to the user program upon I/O completion
  • A wait instruction idles the CPU or a wait loop arises
  • I/O requests are processed sequentially
• The system call requests the OS to wait for I/O completion
  • A device status table contains entries showing I/O device type, address, and status
  • The OS indexes the I/O device table to modify the entry for the table status to include "interrupt"

Storage Definitions and Notation Review

• The base computer storage is a bit which contains 0 or 1
• A byte is 8 bits and is the smallest chunk for storage
• A word is a unit of data size
• Operations execute based on word size, rather than a byte
• Storage is measured in bytes
• A Kilobyte (KB) is 1,024 bytes
• A Megabyte (MB) is 1024^2 bytes = 1,048,576 bytes = 1024 Kilobytes
• A Gigabyte (GB) is 1024^3 bytes = 1,073,741,824 bytes = 1024 Megabytes
• A Terabyte (TB) is 1,024^4 bytes
• A Petabyte (PB) is 1,024^5 bytes
• Computer manufacturers may round off
• Networking uses bits because networks move data bit by bit

Storage Structure

• Main memory is directly accessible by CPU, like RAM
  • It is often volatile
• Secondary storage extends main memory, providing nonvolatile storage
• Hard disks use magnetic recording material with divided tracks and sectors
• A disc controller interacts logical interactions between device (CPU and main memory)
• Solid-state disks are nonvolatile and becoming more popular

Storage Hierarchy

• Storage systems are organized by speed, cost, and volatility
• Caching involves copying information into faster storage and can view main memory as the cache
• Device drivers manage I/O with a uniform interface between controller and kernel

Caching

• Caching transfers information from slower to faster storage temporarily
• Faster storage caches check information
• If it is not there, data copies to cache
• Cache is smaller than the information stored and cache management design is an important consideration

Direct Memory Access (DMA) Structure

• DMA structure uses high-speed I/O devices to transmit data at memory speeds
• It transfers data blocks bypassing CPU intervention so it uses one interrupt per block

Computer-System Architecture

• Most systems use general-purpose processors, also using special-purpose processors
• Multiprocessor systems, are also known as parallel/tightly-coupled systems, growing importance
  • Increased throughput
  • Economy of scale
  • Increased reliability
• Multiprocessor systems have 2 types
• Asymmetric Multiprocessing assigns tasks to different CPUs
  • eg one CPU handles the OS while another performs I/O
• Symmetric Multiprocessing assigns all processor all tasks
  • CPUs all work on the same OS and memory

Dual-Core & Multi-Core Design

• Multi-chip or multicore refers to computers or processors having more than one processing unit (CPU) core

Clustered Systems

• Clustered systems work like multiprocessor systems
  • Share storage via a storage-area network (SAN)
  • Provide a high-availability service for failure survival
• Asymmetric clustering uses one machine in hot-standby node to monitor the system while other nodes do real tasks
• Symmetric clustering involves all nodes running applications and monitoring each other
• HPC enables parallelization
• Distributed lock managers (DLM) prevent multiple operations by maintaining a list to monitor the system

Operating System Structure

• Multiprogramming is needed for efficiency
• A single user cannot keep CPUs and I/O busy -Organizes code and data of the current job
• Total jobs in system are also kept in memory
• Time sharing (multitasking) is a logical extension of multiprogramming It is interactive computing
  • Response time is sub second -Each user can execute one process
    • CPU scheduling if there are multiple ready jobs and the swapping handles when there is no more memory
• Virtual memory allows execution of processes that are not completely in memory

Operating-System Operations

• Operating system operations are triggered by software and hardware interrupts
• A software interrupt can also describe a trap, or exception (divide by zero), process issues - infinite loops

Operating-System Operations (cont.)

• Dual-mode operation protects the OS by switching between user mode and kernel mode
  • A mode bit distinguishes current code execution
  • Privileged instructions only exist in kernel mode
  • System calls switch operations from user to kernel mode
• Increasingly, CPUs such as VMs support multiple operations

Transition from User to Kernel Mode

• Protection of the OS relies on timers to prevent loops that hog resources.
• Timers interrupt the computer after some period
• The OS sets a decrementing counter (privileged instruction)
• The interrupt will generate when zero is reached
• The timer sets before scheduling to control and terminate allotted time