chp 1: intro to operating systems

what is an operating system?

a software that manages a computer’s hardware

what are the goals of an operating system?

• to execute user programs and make solving user problems easier

• make the computer system convenient to use

• use the computer hardware in an efficient manner

what is a kernel

a program running at all times

what are system programs

OS-related but not part of the kernel

what are application programs

everything that are not system programs

what were the 1st generation of computers

dedicated computer; only one program ran at a time and there were no programming languages, so no OS

what were the 2nd generation of computers

shared computers and batch systems — programs were encoded on punch cards (OS/360) is an example

what were the 3rd generation of computers

tape drives and sequentially and multiprogramming was introduced

what were the 4th generation of computers

OSes and embedded systems

what are the 4 components of a computer system

hardware, OS, application programs, and users

what is the user view of an OS

cares about ease of use and good performance + security

what is the system view of an OS

sees the OS as a resource allocator and a control program

what is an interrupt

when device controllers signal the CPU when finished a task by causing an interrupt. CPU stops what it was doing and saves its state (program counter and registers) → jumps to a fixed location and runs service routine) → resumes its task

what is an interrupt vector

table of addresses for the service routines

what is a trap/exception

a software generated interrupt caused by a user error or request

what are the two methods of I/O handling

1. Returning to the user program after the I/O completes

2. Don’t wait for the I/O to complete

• System call: request OS service to wait for I/O

• Device-status table: contains entry for each I/O device: type, address, state

what are the 3 different storages

1. Bootstrap program which is in ROM

2. Volatile storage → so RAM, Cache, etc

3. Nonvolatile storage —> HDD, SDD

what is caching

copying data into faster storage

what is cache coherency

ensuring that the most recent value is in the cache

what is a device driver

provides a uniform interface between the controller and the kernel

what is Direct Memory Access (DMA)

• used for high-speed I/O to transmit into near memory speeds

• device controller transfers data blocks directly into memory — no CPU intervention

• one interrupt per block (not per byte)

what is a core

a component that executes instructions and registers for storing data locally

what is a single processor

one general-purpose CPU; can also have special-purpose CPU’s like the GPU

what are the advantages of a multiprocessor (AKA parallel/tightly coupled)

1. Increased throughput

2. economy of scale

3. increased reliability — graceful degradation or fault tolerance

what are the types of multiprocessing

asymmetric and symmetric multiprocessing

what is asymmetric processing

a master CPU assigns tasks and the slave CPUs can only perform those specific tasks

what is symmetric processing

each CPU can do all tasks and have a shared OS view

what is Non-Uniform Memory Access (numa)

scales by reducing memory contention; common in servers/high-performance systems

what are clustered systems

multiple systems working together, sharing storage via a storage area network (SAN) and provides high availability. some use a distributed lock manager (DLM) to avoid conflicting operations

what is asymmetric clustering

one machine is in hot standby mode

what is symmetric clustering

multiple nodes run applications and monitor each other

what are system daemons

background servies

what is multiprogramming (batch)

keeps subset of jobs in memory, when one waits, the OS switches to another. so it is always running a job

what is multitasking (time-sharing)

a logical extension of multiprogramming where the CPU switches frequently so users can interact. response time should ideally be < 1 sec. swapping and virtual memory help when processes don’t fit fully in memory

what is the reason for a dual mode operation

to protect the OS and other system components

what are the two dual operation modes

user mode and kernel mode. controlled by a hardware mode bit . a system call switches to kernel mode; return switches back to user mode

what are privileged instructions

instructions that only run in kernel mode

what is the point of a timer

prevents infinite loops and hogging resources. the OS sets a timer (which is a privileged instruction) → when it hits 0 → generates interrupt → the OS regains control

what are the 4 pillars of resource management

1. process management

2. memory management

3. file-system management

4. storage management

what is process management

crete/delete process, suspend/resume processes, synchronization, communication, and deadlock handling

what is memory management

decides what is in memory and when; tracks usage, moves data and processes, and allocates/deallocates memory

what is file-system management

provides a logical view of storage, directories, access control, creates and deletes files and maps to disk and backs up

what is storage management

disks store long-term data; OS handles free space, allocation scheduling, positioning, mounting/unmounting

what is protection

controlling access of processes and users to resources

what is security

defending against attacks

what is virtualization

allows OSes to run applications with other OSes; uses VMM

what is emulation

is used when CPU types are different; much slower

what are the two types of client-server environments

1. computer-server system: provides the interface to client and request services

2. file-server system: provides interface for clients to store + retrieve files

what is peer-to-peer environments

does not distinguish clients and servers; all nodes are considered peers

what are cloud computing environments

delivers computing, storage, and apps as a service across a network;

a logical extension of virtualization because it uses virtualization as the base for its functionality

what are the types of cloud computing

public cloud: available via internet to anyone willing to pay

private cloud: run by a company for the company’s own use

hybrid cloud: includes both public and privates cloud componensts

what are real-time embedded systems

the most prevalent form of computers

what connects all the devices

a system bus