sem 4 Flashcards
what are I/O devices?
I/O devices interact with the computer system and the external environment.
input: camera, mic
output: monitor, projector, screen
They key differences the devices have are the data rate, application, complexity of control unit
of transfer, data representation, and error condition.
What are the 3 techniques for performing I/O:
- Programmed I/O
- Interrupt-driven I/O
- Direct Memory Access (DMA)
Programmed I/O:
CPU issues an I/O command to an I/O module and process will be busy waiting for operation to be completed.
not efficient, because I/O process needs to be completed before the process can proceed.
(one example is when we are outputting information to printer to print copies)
Interrupt-driven I/O:
- employ interrupts
- increase efficiency because CPU does not need to wait for the I/O operation to complete before proceeding.
Direct-Memory Access (DMA)
- a DMA module controls the exchange of data between the I/O module and the main memory
– The CPU sends a request to the DMA module, and the CPU will only be interrupted after the entire block has been
transferred
– greatly increases the efficiency (CPU is only involved at the beginning and at the end of the transfer)
what are disks?
Disk is one of the most important devices that
computer communicates with
Both traditional disk and Solid-state Disk (SSD) are
widely used in computers
transfer time equation
– T = b/rN
T is transfer time,
b is number of bytes to be transferred
N is number of bytes on a track
r is rotational speed (rps)
What is RAID
- RAID is an example of fault tolerance.
- RAID uses redundant disk capacity to store parity so
that it can rebuild the data if there is a disk failure. - RAID is widely implemented in servers (as hard disks are bound to failure if running long enough)
How many RAID levels are there?
7 RAID levels
What is RAID Level 0
- have no fault tolerance
- improves the efficiency of the system (performance is much higher)
file is distributed among multiple disks, broken up into multiple pieces, called stripping.
for example, file A can be broken up and located in strip 1, strip 2 and strip 3
What is RAID level 1
- mirror
- duplicates two sets of data
helps to ensure we can recover the data if lost.
can recover minimum 1 disk, maximum 4 if the file size if 4 disks. can recover 4 disks if only in the event 4 disks fails.
What is RAID Level 4?
- uses one disk to store the parity disks
- allows up to one disk failure
What is RAID Level 5
similar to RAID 4 but the parity is striped across all
disks instead of using one whole disk for parity
What is RAID Level 6?
uses two different parity calculations, and the parity
is stored across the different disks
allow up to two disks failure
What is ZFS File system?
RAID cannot protect against data corruption.
Z File System (ZFS) uses checksum to preserve
integrity of data and files
ZFS also provides volume manager, and it can provide
different RAID levels
ZFS is open-source (supported by many Linux
systems)