1.2.1 Systems software - Gurjot

a collection of programs that work together in order to provide an interface between the user and the computer

• memory management • file management • utility software • user interface

when memory is split up into equal sized sections (called pages). these can be swapped between main memory and hard disk as needed.

splitting up memory into logical sized divisions (called segment).

using a section of hard drive to act as RAM when the space in the main memory in insufficient. sections currently not in use are moved into virtual memory through paging

it causes disk thrashing (when the computer freezes as pages are being swapped too frequently between hard disk and main memory.)

signals generated by software or hardware to indicate to the processor that a process needs attention

• Interrupt Request: Device signals the CPU • Acknowledgement and Save Content: CPU finishes current task, saves state to the stack • Identify and Execute ISR: CPU identifies interrupt source and runs ISR • Restore and Resume: CPU restore saved context and resumes main program

jobs are actively made to start and stop by the operating system

• multi level feedback queues • shortest time remaining • round robin

once a job is started, its left alone until completed

• first come first served • shortest job first

each job is given a section of processor time, if not completed in the time, its put back in the queue.

jobs are processed in order of which entered the queue first

makes uses of multiple queues, ordered based on different priority.

jobs processed in ordered according to the time required for completion, with the shortest at the front

the queue storing jobs is ordered according for the time left for completion. if a new job enters the queue, its compared to the one being processed and is replaced with the one being replaced if it has a shorter completion time.

when a process does not receive enough time in order to be completed.

• shortest time remaining • shortest job first

an os run across multiple devices, so load is spread across multiple processors.

built to perform a small range of specific tasks catered to a specific device

enables user to carry out tasks seemingly simultaneously using time slicing

makes use of a supercomputer , and scheduling algorithm to ensure time is shared fairly between jobs

designed to perform a task in a guaranteed time frame