1 Lecture 11: Computer Systems and Networks ITEC 1000 “Introduction to Information Technology”

1

Lecture 11:Lecture 11:Computer Systems and NetworksComputer Systems and Networks

ITEC 1000 “Introduction to Information Technology”

2

Lecture Template:Lecture Template:

System Performance ImprovementSystem Performance Improvement MultiprocessingMultiprocessing Computer InterconnectionComputer Interconnection OSI model and TCP/IPOSI model and TCP/IP Network TopologiesNetwork Topologies Protocols Protocols Wide Area NetworkWide Area Network High Performance ComputingHigh Performance Computing

3

System Performance System Performance ImprovementsImprovements

Multiple CPUsSharing memory and I/O facilitiesConflict among the CPUs for shared resources

Faster clock speed, buses and circuitsImproving technology to design faster CPU circuits and buses

Wider instruction and data pathsWider interface between the CPU and memory allows to fetch more data in a single operation

Faster disk accessSmaller discs, more density packed, increased storageRAID: data on different discs, multiple access simultaneously

More and faster memoryReduces the time to access instructions and data

4

MultiprocessingMultiprocessing

ReasonsIncrease the processing power of a systemParallel processing

Types of multiprocessor systemsTightly coupled systemsLoosely coupled systems

5

Tightly Coupled SystemsTightly Coupled Systems

Also called multiprocessor systems Identical access to programs, data,

shared memory, I/O, etc. Easily extends multi-tasking, and

redundant program execution Two ways to configure

Master-slave multiprocessingSymmetrical multiprocessing (SMP)

6

Tightly Coupled SystemsTightly Coupled Systems

7

Master-Slave MultiprocessingMaster-Slave Multiprocessing

Master CPUManages the systemControls all resources and schedulingAssigns tasks to slave CPUs

AdvantagesSimplicityProtection of system and data

DisadvantagesMaster CPU becomes a bottleneckReliability issues – if master CPU fails entire system fails

8

Symmetrical MultiprocessingSymmetrical Multiprocessing

Each CPU has equal access to resources Each CPU determines what to run using a

standard algorithm Disadvantages

Resource conflicts – memory, i/o, etc.Complex implementation

AdvantagesHigh reliabilityFault tolerant support is straightforwardBalanced workload

9

Loosely Coupled SystemsLoosely Coupled Systems

Clusters or multi-computer systems Each system has its own CPU, memory, and

I/O facilities Each system is known as a node of the

cluster Advantages

Fault-tolerant, scalable, well balanced, distance is not an issue

Two ways to configureShared-nothing modelShared-disk model

10

Shared-Nothing ModelShared-Nothing Model

High speed link between nodes No sharing of resources Partitioning of work through division

of data Advantage

Reduced communication between nodes Disadvantage

Can result in inefficient division of work

11

Shared-Disk ModelShared-Disk Model

High speed link between nodes Disk drives are shared between nodes Advantage

Better load balancing Disadvantage

Complex software required for transactional processing (lock, commit phases)

12

Cluster ModelsCluster Models

13

Computer InterconnectionComputer Interconnection

Communication channel – pathway for data movement between computers

Point-to-Point connectivityCommunication channel that passes data directly between two computersSerial connectionTelephone modemTerminal controller – handles multiple point-to-point connections for a host computer

Multipoint connectivityMultidrop channel or shared communication channel

14

Example: Point-to-PointExample: Point-to-Point

15

DefinitionsDefinitions

Topology: the way in which loosely coupled computers are interconnected

Synonym: configuration

Protocol: a set of rules and standards for communications between computers

16

Client-Server ArchitectureClient-Server Architecture

Computer server provides servicesFile storage, databases, printing services, login services, web services

Client computersExecute programs in its own memoryAccess files either locally or can request files from a server

17

Client-Server NetworkClient-Server Network

F

server

18

A typical data packetA typical data packet

19

The layers of the OSI modelThe layers of the OSI model

20

Passing a message through an Passing a message through an intermediate nodeintermediate node

21

A comparison of OSI and TCP/IPA comparison of OSI and TCP/IP

22

Beowulf ClustersBeowulf Clusters

Simple and highly configurable Low cost Networked

Computers connected to one another by a private Ethernet networkConnection to an external network is through a single gateway computer

ConfigurationCOTS – Commodity-off-the-shelf components such as inexpensive computersBlade components – computers mounted on a motherboard that are plugged into connectors on a rackEither shared-disk or shared-nothing model

23

Blade and Rack of Beowulf ClusterBlade and Rack of Beowulf Cluster

24

LAN TopologyLAN Topology

Arrangement of workstations in a shared medium environment

Logical arrangement (data flow) Physical arrangement (cabling

scheme)

25

Network TopologiesNetwork Topologies

Bus: Nodes are connected to a common bus with a terminator on each end.

Ring: Similar to bus, but bus is closed - no ends.Star: A central node does most of the processing. Remote

nodes are connected point-to-point with it.Loop: Similar to ring but nodes are directly in

communication path.Hierarchical: A tree-like structure emanating from a central or

root node.Web: Everything connected to everything else.

26

LAN Topologies: RingLAN Topologies: Ring

Repeaters are joined by unidirectional point-to-point links in a ring

As data circulates past a receiver, the receiver checks its address, and copies those intended for it into a local buffer

Data circulates until it returns to source, which removes it from network

Better performance at high levels of usage

27

Ring LAN DiagramRing LAN Diagram

28

Ring TopologyRing Topology

29

LAN Topologies: Bus

Multipoint medium Stations attach to linear medium (bus)

using tap Transmission from any stations

travels entire medium (both directions)

Termination required at ends of bus to prevent the signal from bouncing

Break in cable brings down entire bus

30

Bus LAN DiagramBus LAN Diagram

31

Bus TopologyBus Topology

32

LAN Topologies: TreeLAN Topologies: Tree

Generalization of bus topology Branching cable with no closed loops Cable(s) begin at headend, travel to

branches which may have branches of their own

Each transmission propagates through network, can be received by any station

33

LAN Topologies: StarLAN Topologies: Star

Each station connected point-to-point to a central station, usually with two unidirectional links

Switching in the central station connects pairs of nodes together

Central node can broadcast info, or can switch frames among stations

Failure of central station causes entire network to go down

34

Star LAN DiagramStar LAN Diagram

35

Star TopologyStar Topology

Guy inthe

Middle

36

Loop TopologyLoop Topology

37

Hierarchical TopologyHierarchical Topology

Big Guy at Top

38

Web TopologyWeb Topology

39

Topology Selection: FactorsTopology Selection: Factors

Distances between stations Layout of the room/building Overall size of the network Distance between the most remote

nodes Speed requirements Network traffic Total number of stations

40

Medium Access Control (MAC) Medium Access Control (MAC) ProtocolsProtocols

Characteristics of the channels, data rate, voltage levels, etc.

Node access to the channel (medium access control protocol)

Steer data to its destinationDetect errorsPrevent multiple nodes from accessing the network simultaneously (collision)Ethernet and token ringImplemented in hardware

41

Ethernet MAC ProtocolEthernet MAC Protocol

MAC – Medium Access Control Ethernet and CSMA/CD

Carrier sense multiple access with collision detection

Four step procedureIf medium is idle, transmitIf medium is busy, listen until idle and then transmitIf collision is detected, cease transmittingAfter a collision, wait a random amount of time before retransmitting

42

Ethernet FrameEthernet Frame

43

Switched EthernetSwitched Ethernet

44

Token Ring MAC ProtocolToken Ring MAC Protocol

Token “seized” by changing a bit on the circulating frame to indicate start of frame rather than token

Default configuration requires sender to complete transmission and begin receiving transmitted frame before releasing the token

“Early token release” allows release of token after transmission but before receipt of frame

45

HubsHubs

The active central element of the star layout

When a single station transmits, the hub repeats the signal on the outgoing line to each station

Hubs can be cascaded in a hierarchical configuration

Ethernet hubs are physically a star but logically a bus

46

BridgesBridges

Allow connections between LANs and to WANs

Used between similar networks Read all frames from each networkAccept frames from sender on one network that are addressed to a receiver on the other networkRetransmit frames from sender using MAC protocol for receiver

47

GatewaysGateways

Similar to bridges but connect dissimilar networks

Convert format of the message to correspond to the protocol of the other network

Network traffic is specifically addressed to the router

48

Wide Area NetworkWide Area Network

Circuit switchingDedicated channel between source and destination for duration of connection

Message switchingDedicated channel for an entire message

Packet switchingAn independent path is created for each datagram

Virtual circuit switchingA route is created from source to destination before transmission begins and all datagrams are sent using the same route

49

A Switched Wide Area NetworkA Switched Wide Area Network

50

Networks vs. ClustersNetworks vs. Clusters

Externally, clusters appear as a single computing unit

Network nodes are individually identifiable

Workload on a cluster is determined by cluster administration and load-balancing software

Network workload cannot be controlled using the above method

51

High Performance ComputingHigh Performance Computing

Massively parallel processor architectures (MPP)

Clusters of power machines or larger Beowulf blade clusters

Well suited for problems that can be broken into subtasks

Grid computingSupercomputer performance through distributing CPU processing to the spare CPU cycles of personal computers connected to a network

52

Parallel ComputersParallel Computers

Massively parallel architectures Hundreds to millions of CPUs CPUs have small amounts of local memory All CPUs have access to global shared

memory Pipelined CPUs

Results from one CPU flow to the next CPU for additional processing