CNIT 176 - Lecture notes All PDF

Preview

Summary

Notes and Quiz answers for the course...

Description

CNIT 176 WTHR 104 HEAV B008 NM Hands [email protected]

System Architecture A system is a collection of components linked together in such a way so that they can be recognized as one.

The environment is anything outside of the system

The interaction of system with the environment is called the architecture of the system

Client Server Computer A program on a client computer requests services from a host computer eg. emails, web page lookups. Centralization of services permits: easier administration consistency of resources such as files and data can be managed and assured more efficient and cost effective hardware

Types of data format collating sequence Why UTF-8

Data Encoding EBCDIC extended binary coded decimal interchange code, limited to IBM mainframes common in archival data character codes differ from ASCII

ASCII 7 bit telecommunications code. One of the most dominant until 10 years ago. LSD - least significant digit (right side) MSD - Most significant digit (left side)

Unicode what we use today 16 bit can be called a superset of ASCII allows software modification for local languages UTF-8 is an implementation of Unicode , one of the most popular implementation of Unicode.

Non-Text Data Encoding Compression - recoding data so that it takes lesser space lossless: can restore the data to original state lossy: original form of data cannot be retrieved Compression Ratio - amount of file size reduced

Visual Data Raster or Bitmap Used for realistic images with continuous variations. Each pixel in a graphic is stored as a binary number Vector or Object

Boolean Algebra Truth table example

Reducing boolean algebra Signed mgnitude the MSD is used to store the sign bit (+ or-) 0 = positive number 1 = negative number E.g. -1 would be 10000001 Signed magnitude - overflow This is when two signed bits are added togther an overflow causes an incorrect result

Complement Systems allows subtraction 10-7 = 10+(-7) this simplifies the notion so that it can be added in binary 1's complement = Positive binary number > flip the bits 2's complement = 2 d −

N or take a number > flip it and add 1

Registers Small permanent storage locations Manipulated by the control unit CU size in bits or bytes can hold data, address or instruction

Memory has a unique address - a byte is the smallest addressable unit address in an instruction is copied to the MAR CPU decides if the operation is storage or retrieval

Transfer takes place between MDR and memory MDR is a two way register

Memory switch determines whether we are writing or reading to a cell. Memory capacity is determined by two things 1. Number of bits in MAR a.

2k where k is the width of the register in bits

2. Size of address portion of the instruction a. 4 bits allow 16 locations (24 ) b. 34 bits allow ~ 4 GB c. 64 bits allow 16 billion gigabytes Therefore, a 32 bit computer can only make use of 4 GB of RAM

RAM DRAM Volatile - does not store a state SRAM

Non-volatile - can store a state Faster than DRAM SD RAM (Synchronous Data RAM) SDR (Single data RAM) DDR (Double data RAM)

Fetch Decode Execute Cycle Fetch find the instruction , load from memory into register and ALU Execute Perform a set of instruction Move data Loading Data:

Storing Data:

Add Data:

Virtual Memory Secondary memory that a computer uses as a primary storage HDD acting as RAM A chunk of virtual memory is called a page The area on your hard disk that is storing the instructions that were previously in RAM is referred to as a page file The page frame is the physical chunk in RAM the virtual page is mapped to. When the RAM is full i.e.. all the pages are full, the RAM swaps a new page for an old one with the virtual memory of the secondary storage

Overloading this process is called Thrashing During thrashing the CPU sits idle while the RAM is continuously swapping pages with the virtual memory

BUS physical connection that makes data transfer possible electrical or optical conductors parallel - info can travel at the same time requires a lot of space more data travels at a time used for short distances eg CPU serial - info travels one by one eg USB longer distances

Simplex - unidirectional full duplex - bidirectional half duplex - bidirectional but one directional at one time

Shifting

info in the bit bucket is lost commands jump branch subroutine call return

Stacks and Queues stacks LIFO - last in first out PUSH - add data to the stack POP - remove data from the stack

stack pointer - location of the stack Queues FIFO - first in first out

Networking Packets chunks of data containing something known as a packet header which contains information about the packet packets allows multiple nodes to be connected and a channel can be shared. provides much more effective communication Also known as frame, or datagram ethernet packets are called frames streaming data packets are called datagrams

Packet Header contains logical address and different protocols ports sequence information size of data etc.

Channel Characteristics communication channel path between two nodes on a network may also include intermediate nodes that forward packets Links segment of a communication channel Bandwidth Data transfer rate ( bits per seconds) Medium

Guided- communication is limited to a specific path e.g. on a wire Unguided - communication is not limited to a specific path e.g. wifi Direction simplex, half duplex, full duplex Number of connections point to point or multipoint End node interface wired or wireless BT, DSL, cable etc.

Network Topology layout of the network logical and physical topology 1. Mesh Topology - Multiple nodes between networks

2. Bus Topology - The message is broadcasted to every node on the network

3. Star Topology - A central switch controls the flow of the messages (most common today) e.g. LAN

4. Tree Topology - Combination of star and bus topology e.g. DNS!

5. Ring Topology - requires lots of switches (expensive to implement)

Local Area Network connects computers over small local area Uses ethernet standards determined by IEEE

Ethernet Hub based on bus topology Broadcasts the message Ethernet Switches based on star topology direct connection between two nodes

Hub vs Switch

Backbone Network connects LAN together Image

Metropolitan Area Network

larger geographical area than LAN range of less than 30 miles or 50 km Image

Wide Area Network (WAN) Large geographical area (globally) surf-net

Piconet / Personal Area Network (PAN) for individual use range of 30 feet or less e.g. Bluetooth

Packet Routing circuit switching old fashioned telephone line (landline) virtual circuit multi link channel path there are intermediate nodes packet switching Packet routed note to node independently taking different routes

Router & Gateway A switch is used to create a network and a router is used to interconnect networks. Gateway connects dissimilar networks together e.g. analog to digital channels, networks with different protocols etc.

TCP/IP Model model based on protocols

physical layer here is different than the one in osi model

Physical Layer actual communication taking place consisting of bare stream of bits

Data Link Layer

Encapsulation and De-encapsulation

Bridges Device used to connect two or more hosts together bridging connects two separate networks as if they were a single network

OSI Model A largely theoretical models...