Systems Exploitation: Overview

• Systems exploitation is covered in chapter 1 of the Generalities section of the study notes
• The target audience is L1 TI
• The teachers are Bechir HAMIDI
• The version is 2025

Plan Overview

• The study notes will describe an overview of systems exploitation, and its historical role, evolution and core
• It will also describe the file management system with architecture and its organisation
• Will discuss the management of processes and threads, including process descriptions, the core and sequencing
• It will analyse memory management and E/S management
• It will cover the fundamentals of embedded systems.

Plan Summary

• The summaries includes definitions
• The evolution of the systems exploitation will be covered
• The principals and components of components of a system exploitation are described
• The classification of systems will be covered

Operating System (OS) Definition

• An OS is a fundamental program that manages computer resources and provides a user interface for hardware interaction
• It controls input/output devices, memory, the processor, and other computer components
• The purpose includes increasing resource availability for users
• and optimizing hardware use

Common OS Examples

• Windows is the most used operating system on personal computers. (Windows 10/11)
• Linux is an open-source operating system widely used in servers, embedded systems, and development environments
• MacOS is Apple's operating system, based on Unix, known for its user interface and integration with the Apple ecosystem
• Android/iOS are the dominant mobile operating systems. A specific architecture manages limited resources (memory, battery)
• Embedded systems such as FreeRTOS or QNX, are used in IoT devices and self-driving cars

Technological Evolution (1945-1955)

• The first generation of computers, covered between 1945 and 1955, did not have an OS to share resources
• Only one program benefitted from all of the machine's resources
• User interaction was direct, where users executed step-by-step instructions and could directly change memory content without restriction

Technological Evolution (1955-1965)

• The second generation of computers operated between 1955-1965, and did Batch processing using punch cards
• The sequence of lots was described by a special punch card (control card)
• The OS was limited to a resident monitor that chained the lots (chain monitors).

Technological Evolution (1965-1970)

• The third generation of systems evolved between 1965 and 1970, introducing key modern OS concepts: multiprogramming, time-sharing, and virtual memory.
• Multiprogrammed systems included multiple simultaneous tasks in memory
• It allowed for multiplexing of the processor between tasks
• There was some processor loss during I/O, however
• The core advantage was it enabled better use of the CPU
• Disadvantages include complexity, large memory capacity
• It also requires resouce sharing and resource protection

Shared time systems (1970)

• Shared time systems involve division of processor time between tasks
• It introduces the concept of Quantums of time
• Tasks can be resident in memory or "swapped" on disk
• More generally, these systems enabled a greater number of tasks
• Allocated more memory to each task
• Advantages- Correct response times for short tasks, especially in the presence of long, non-interactive tasks
• A major Inconvenient is that complexity and processor usage is may be less that optimal

Fourth Generation Evolution (After 1970)

• The introduction of personal computers marked the start of this era
• Specialty architectures, such as Parallel, Real-Time and Embedded systems became prolific
• Evolution moved toward to distributed systems using cluster or cloud based architectures
• There was common quality criteria such as: Reliability, availability, efficiency, extensibility, adapted

Cloud OS Evolution

• Cloud computing has led to operating systems in the cloud
• It enables users to access applications and data from any device with an Internet connection.

Principal OS Functions

• Processor management: manages allocation of the processor among programs. (process management).
• Active memory management: manages memory space for each app and can create memory on the hard disk (virtual memory).
• Input/output management handles information exchange between the comp and peripheral devices
• Rights management is responsible for security related to program execution
• File management handles file system reading/writing and file access rights

OS Components

• The kernel represents the core OS functions for managing memory, processes, main input/output, etc.
• The command interpreter facilitates communication with the system through a command language
• The file system is organized to enable registering files in a tree-like structure

System Classification

• Multitasking systems execute multiple processes at once
• For instance, compiling and consulting the source code file
• This classification includes UNIX, OS/2 by IBM, and Windows 95
• Multiuser systems have multiple simultaneous users sharing the same hardware
• UNIX, MVS, and Gecos are examples of this
• Centralized systems are entirely present on a single machine
• Connected machines act as separate entities from the machine with the centralized system
• The system only manages resources on local machines
• Applications such as X11, FTP, and Mail are developed using UNIX

Distributed Systems

• Distributed systems feature components spread across multiple machines (sites)
• To users, a distributed operating system appears as a single virtual uniprocessor, even when that is not the configuration
• Users are not aware of resource locations and do not need to know which machine executes a program

Real-time Systems

• Hard Real-Time Systems: Typically have little to no secondary memory
• They do not share time
• They feature a short switching time
• They are specialized and context dependent
• Soft Real-Time Systems are usually used in multimedia, and virtual reality applications
• These systems have constraints for more flexible time
• Task priorities assigned
• Specific memory management techniques are used

Mobile Systems

• Mobile systems are typified by smart phones
• Require specialized operating systems
• System constraints include limited memory, slow processors, and small screens
• They generally have low energy consumption
• Have high responsiveness

Parallel Multiprocessor Systems (SMP)

• Symmetric Multi Processors are examples of parallel multiprocessor systems
• These include machines with parallel computing capability
• They are generally highly specialized architectures

Clustered Systems

• These systems are also known as server clusters, or computer farms
• Systems divided across multiple nodes
• Such as Homogeneous systems
• The systems either share a disk, or share nothing
• They are typically connected by a specialized network
• These include SCI, InfiniBand, and Myrinet
• Enable rapid exchanges of large chunks of data
• Three characteristics- Able to support a higher load without being restructured, allows horizontal scaling, and has high availability using backup servers

Computing Grids

• Computing Grids are typically distributed systems
• They are distributed infrastructures which are composed of many different machines, and may be located in different physical locations
• Example: Grid5000 uses around 3000 processors across 9 sites in France.
• Characterized by intensive computation
• Designed with Guaranteed availability
• And feature a material resources reservation system

Cloud Computing

• Cloud Computing: A shared system where clients utilize services from remote servers without managing the underlying infrastructure
• Cloud delivered services can run within cluster architectures
• Three main models
  • Infrastructure as a Service (IaaS)
  • Platform as a Service (PaaS)
  • Software as a Service (SaaS)

OS Internal Structures

• Layered Structure: OS organized in layers where each layer uses functions of the lower layers. Difficulty lies in defining the different layers.
• Monolithic Structure: Procedures can call each other at any time. Organized in 3 layers:
  • A main procedure in the upper layer identifies required service,
  • Service procedures in the layer below execute system calls
  • Utility procedures in the lower layer aid system procedures and can be called by multiple procedures.
• Virtual Machines: The core handles multiprogramming, giving the layer above multiple virtual machines. Different OS can run on a virtual machine.
• Client/Server Model: The OS comprises a kernel and a set of servers. The kernel manages communication between clients and servers. Clients request services.