COSC3360 EXAM1

basic elements of OS

processor, I/O, main memory, system bus

what does processor do

this 1) controls the operation of the computer and 2) performs data processing functions

processor aka

CPU aka

what does main mem do

this stores data and programs

what is the one thing about main mem when a comp is shut down

when comp is shut down, contents of mem is lost

main mem aka

real mem/ primary mem aka

what do I/O modules do

they move data btwn comp and its external environment

what does system bus do

this provides for comms among processors, main mem and I/O modules

what is microprocessor

it is a processor on a single chipwh

what is GPU

this is graphical processing unit

what does GPU do

this provides efficient computation on data using SIMD techniques

GPU usage is found in (something really common these days)

AI uses this kind of unit

what is DSP

this is digital signal processor

what does DSP do

this deals w streaming signals like audio or vid, encoding/ decoding speech and vid, provide support for encryption and security

DSP is now (found in handheld devices)

this is now first-class computational device

where is set of instructions stored

a set of instructions stored in mem

two steps to executing instruction

the steps to instr execution are 1) processor fetches instr from mem and 2) processor executes

what is function of program counter

the func of this is to hold address of the next instr to be fetched

what happens to pc after each fetch

after each fetch, this is incremented

where does fetched instr go

this goes or gets loaded into instruction register IR

what are interrupts

they affect the normal sequencing of processor

goal of interrupt

the goal of this is to improve processor utilization

reasons for using interrupts

reasons for using this: most I/O devices slower than processor and therefore processor must wait, resulting in wasteful use of processor

classes of interrupts:

the classes of this are program, timer, I/O, and hardware failure

program interrupt is

this type of interrupt is when an error w instr execution like calculation, illegal instr, or data overflow

time interrupt is

this type of interrupt is generated by a timer within the processor

I/O interrupt is

this type of interrupt is generated by I/O controller/ devices like completion of operation or to signal error

hardware failure interrupt is

this type of interrupt is generated by a failure like power failure or mem error

what happens when hardware interrupts

processor finishes executing current instr, acknowledges interrupt, pushes PSW and PC onto control stack, and loads new PC value

what happens when software interrupts

remainder of process state info is stored, then process interrupt, restore process state info, and restore PSW and PC

two approaches to when an interrupt occurs while another interrupt is being processed

1) disabled interrupts when one is being processed or 2) priority scheme

faster access time in mem means

faster access time in this means greater cost

greater mem capacity

large capacity of this means smaller cost and slower access speed

going down the mem hierarchy

decreasing cost, increasing capacity, increasing access time, decreasing frequency of access to mem by processor

principle of locality is

this principle is how close processor is to mem

what is secondary mem

this aka auxillary mem or external mem

what does secondary mem store

this type of mem stores program and data files

characteristics of cache mem

this type of mem is invisible to OS

interacts w other mem management hardware

exploits principle of locality with a small and fast mem

cache mem regarding processor

this type of mem, processor must access it at least once per instr cycle and processor execution is limited by mem cycle time

smaller caches have … impact on performance

this type of mem has significant impact on performance

what is block size

this is unit of data exchanged btwn cach and main mem

characteristics of cache mem

cache size, block size, number of cache levels, write polity, mapping function and replacement algorithm

what is mapping function in cache mem

this func in cache mem determines which cache location the block will occupy

how does cache replacement algorithm work

this algorithm works by choosing which block to replace when a new block is to be loaded into cache using LRU (least recently used) technique

what is write policy in cache mem

this in cache mem determines when mem write operation takes place

when can mem write operation occur

this operation can happen every time block is updated or replaced

when the processor sees an instr relating to I/O

when this happens, the processor executes accordingly to the appropriate I/O module

types of techniques to execute instr relating to I/O

programmed, interrupt driven or DMA (direct mem access)

programmed I/O works by

this type of I/O performs the requested action and periodically checks status of I/O module until instr is complete; severely degrades performance level of entire system

interrupt-driven I/O works by

this type of I/O works by processor issuing an I/O command, go on to do other work and when I/O is ready to exchange data w processor, it interrupts the prog, processor then executes that I/O and resumes its former processing

drawback of interrupt-driven I/O

the drawback of this is that transfer rate is limited by speed of which processor can test and service a device

DMA I/O works by

this type of I/O works by processor issuing a command to DMA module when processor wants to read or write data

what does a command to DMA from processor need to have regarding I/O

this command from this to this needs: read/ write request, address of I/O device involved, starting location in mem to read/write and number of works to read/write (this whole process is done by a separate module on system bus or incorporated into I/O module)

one thing ab DMA I/O is

entire block of data is directly transferred to and from mem without going thru processor; processor only involved at beginning and end of transfer

DMA is most efficient compared to interrupt-driven and programmed I/O

DMA is most efficient compared to interrupt-driven and programmed I/O

characteristics of symmetric multiprocessors (SMP)

this kind has two or more similar processors, processors share same main mem and connected by a bus, processors share access to I/O devices, and all processors can perform same functions

pros of SMP

pros of this are: greater perf, at different price point is its according perf, failure of a single processor does not halt machine, additional processor can enhance perf

multicore computer aka

aka chip multiprocessor

what is it about multicore computer

this kind of computer has two or more processors (cores) on a die

each core in multicore comp has

this in multicore comp has all the components of an independent processor

what is operating system

this system controls execution of application progs

ISA stands for

this stands for Instruction set arch

ABI stands for

this stands for app binary interface

API stands for

this stands for app programming interface

the OS is repsonsible for (as a resource manager)

it is responsible for controlling the use of a computer’s resources (as a resource manager)

what are the desirable hardware features

the desirable ___ware features are memory protection, timer, privileged instructions, and interrupts

what are the two modes of operation

the two modes of this are user mode and kernel mode

what happens in user mode of operation

in this mode of operation, certain areas of memory are protected and certain instructions may not be executed

what happens in kernel mode of operation

in this mode of operation, privileged instr may be executed and protected areas of memory may be accessed

what is simple batch system

in this kind of system, processor time alternates between execution of user progs and execution of monitor

what is multiprogrammed batch systems

in this kind of system, processor is often idle

what does uniprogamming wait for

this kind of programming waits for an I/O instr to complete before proceeding

what does multiprogramming mean for processor

this type of prog allows processor to switch btwn jobs that are not waiting for I/O

what is time sharing systems

in this kind of system, processor time is shared among multiple users and multiple users can simultaneously access the system thru terminals w OS interleaving the execution of each user prog

difference between multiprogamming and time sharing

the difference between these two systems are multiprog max processor use and time sharing min response time, multuprog has job control lang provided w the job and time sharing does commands at terminal

what is a fundamental aspect to the structure of OS

process is ____ to the structure of OS