CECS 326 EXAM 3 STUDY

Which algorithm minimizes average waiting time?

FCFS

RR

SJF

Priority

SJF

Round Robin with a very large time quantum behaves like

SJF

FCFS

Priority

SRTF

FCFS

Which scheduling algorithm is preemptive

FCFS

SJF

SRTF

Non-preemptive priority

SRTF

What is the main issue with priority scheduling

Starvation

Deadlock

Fragmentation

Thrashing

Starvation

In multilevel queue scheduling, processes

stay in one queue

share CPU equally

Move between queues

Are random

stay in one queue

Which is directly accessed by CPU

Disk

Cache Only

Virtual Memory

Registers and Main memory

Registers and Main Memory

MMU is responsible for

Disk allocation

Scheduling

Address Translation

Paging

address translation

Best-fit Allocation chooses

First hole

Largest Hole

Smallest suitable hole

Random hole

Smallest suitable hole

External Fragmentation occurs when

Memory too small

Memory not contiguous

CPU overload

Cache Miss

Memory not contiguous

Paging Eliminates

External Fragmentation

Internal Fragmentation

Both

None

External Fragmentation

Virtual memory allows

Programs larger than RAM

only physical memory usage

no disk usage

faster CPU

Programs larger than RAM

Demand Paging loads pages

at compile time

at load time

when needed

after execution

When needed

A page fault occurs when

CPU error

disk failure

Page is missing

Page is in memory

Page is missing

FIFO replaces

random page

least used page

newest page

oldest page

Optimal Algorithm replaces

least used page

most recent page

page not used longest in future

random page

page not used longest in future

Thrashing happens when

Too many page faults

too few process

CPU idle

disk full

too many page faults

a file is

random memory

disk sector

cache unit

logical address space

logical address space

Which attribute is human-readable

name

ID

location

size

Name

Sequential access processes data

randomly

backwards

in order

once

in order

Tree directory structure provides

Unique Paths

One file

No organization

No hierarchy

Unique Paths

Contiguous allocation stores files

Randomly

in linked blocks

continuously

in cache

continuously

Linked Allocation drawback

fast access

slow random access

fragmentation

large memory

slow random access

Big data grows

slower than Moore’s law

same rate

faster than Moore’s law

not growing

faster than Moore’s law

Which is not a Big Data V

Volume

Velocity

Variety

Value

Value

Map function outputs

files

key-value pairs

processes

memory blocks

key-value pairs

Reduce function

splits data

combines values

deletes data

stores data

combine values

MapReduce works by:

Sequential processing

single CPU

cache only

parallel chunks

parallel chunks

Hadoop is designed for

distributed systems

local computing

cache management

CPU scheduling

distributed systems

SJF is always practical in real systems

false

SRTF is preemptive SJF

true

Round Robin always gives best turnaround time

false

Logical and physical addresses are always the sae

false

Paging removes external fragmentation

true

Paging removes internal fragmentation

false

TLB speeds up address translation

true

Swapping moves processes between memory and disk.

true

Virtual memory requires entire program in RAM.

False

Page faults are expensive operations.

true

FIFO can suffer from Belady’s anomaly.

True

LRU uses future knowledge.

False

Single-level directory is scalable.

False

Tree directories are most common.

true

Linked allocation supports fast random access.

false

Contiguous allocation has good performance.

true

Big data includes structured and unstructured data.

True

MapReduce is sequential.

false

Failures are common in distributed systems.

true

Each worker processes the entire dataset.

false