CpE 242 Computer Architecture and Engineering Interconnection NetworksRecap: Advantages of BusesRecap: Disadvantages of BusesRecap: Types of BusesRecap: Increasing the Bus BandwidthBus Summary:Outline of Today’s LectureNetworksCurrent Major NetworksSlide 10ABCs of NetworksA Simple ExampleQuestions about Simple ExampleProtocol StacksInterconnection NetworksInterconnection Network IssuesImplementation IssuesMediaSlide 19Network Performance MeasuresExample Performance MeasuresImportance of Overhead (+ Latency)Slide 23TopologyImportant TopologiesFat TreeConnection based vs. ConnectionlessPacket formatsExample: Ethernet (IEEE 802.3)Example: ATM (Asynchronous Transfer Mode)Towards the Killer NetworkSummary: Interconnectionscs 152 nets.1©DAP & SIK 1995CpE 242Computer Architecture and EngineeringInterconnection Networkscs 152 nets.2©DAP & SIK 1995Recap: Advantages of Buses°Versatility:•New devices can be added easily•Peripherals can be moved between computersystems that use the same bus standard°Low Cost:•A single set of wires is shared in multiple waysMemoryProcessorI/O DeviceI/O DeviceI/O Devicecs 152 nets.3©DAP & SIK 1995Recap: Disadvantages of Buses°It creates a communication bottleneck•The bandwidth of that bus can limit the maximum I/O throughput°The maximum bus speed is largely limited by:•The length of the bus•The number of devices on the bus•The need to support a range of devices with:-Widely varying latencies -Widely varying data transfer ratesMemoryProcessorI/O DeviceI/O DeviceI/O Devicecs 152 nets.4©DAP & SIK 1995Recap: Types of Buses°Processor-Memory Bus (design specific)•Short and high speed•Only need to match the memory system-Maximize memory-to-processor bandwidth•Connects directly to the processor°I/O Bus (industry standard)•Usually is lengthy and slower•Need to match a wide range of I/O devices•Connects to the processor-memory bus or backplane bus°Backplane Bus (industry standard)•Backplane: an interconnection structure within the chassis•Allow processors, memory, and I/O devices to coexist•Cost advantage: one single bus for all componentscs 152 nets.5©DAP & SIK 1995Recap: Increasing the Bus Bandwidth°Separate versus multiplexed address and data lines:•Address and data can be transmitted in one bus cycleif separate address and data lines are available•Cost: (a) more bus lines, (b) increased complexity°Data bus width:•By increasing the width of the data bus, transfers of multiple words require fewer bus cycles•Example: SPARCstation 20’s memory bus is 128 bit wide•Cost: more bus lines°Block transfers:•Allow the bus to transfer multiple words in back-to-back bus cycles•Only one address needs to be sent at the beginning•The bus is not released until the last word is transferred•Cost: (a) increased complexity (b) decreased response time for requestcs 152 nets.6©DAP & SIK 1995Bus Summary:°Bus arbitration schemes:•Daisy chain arbitration: it cannot assure fairness•Centralized parallel arbitration: requires a central arbiter°I/O device notifying the operating system:•Polling: it can waste a lot of processor time•I/O interrupt: similar to exception except it is asynchronous°Delegating I/O responsibility from the CPU•Direct memory access (DMA)•I/O processor (IOP)cs 152 nets.7©DAP & SIK 1995Outline of Today’s Lecture°Recap and Introduction (5 minutes)° Introduction to Buses (15 minutes)°Bus Types and Bus Operation (10 minutes)°Bus Arbitration and How to Design a Bus Arbiter (15 minutes)°Operating System’s Role (15 minutes)°Delegating I/O Responsibility from the CPU (5 minutes)°Summary (5 minutes)cs 152 nets.8©DAP & SIK 1995Networks°Goal: Communication between computers°Eventual Goal: treat collection of computers as if one big computer°Theme: Different computers must agree on many things => Overriding importance of standards°Warning: Buzzword rich environmentcs 152 nets.9©DAP & SIK 1995Current Major NetworksARPA netT1, 56KbpsT3, 230KbpsIP - internet ProtocolTCP - TransmissionControl ProtocolToken RingEthernetFDDICS NetRelayX.25 (Telenet, Uninet_PhonenetCS NetBitnetNSF Net1.6Mbps10 Mbps4Mbps100MbpsATMcs 152 nets.10©DAP & SIK 1995Networks°Facets people talk a lot about•direct vs indirect•topology•routing algorithm•switching•wiring°What matters•latency•bandwidth•cost•reliabilitycs 152 nets.11©DAP & SIK 1995ABCs of Networks°Starting Point: Send bits between 2 computers°FIFO Queue on each end°Can send both ways (“Full Duplex”)°Rules for communication? “protocol”•Inside a computer? •Loads/Stores: Request(Address) & Response (Data)•Need Request & Response•Name for standard group of bits sent: Packetcs 152 nets.12©DAP & SIK 1995A Simple Example°What is format of packet?•Fixed? Number bytes?Request/ResponseAddress/Data1 bit32 bits0: Please send data from Address1: Data corresponding to requestcs 152 nets.13©DAP & SIK 1995Questions about Simple Example°What if more than 2 computers want to communicate?•Need computer address field in packet?°What if packet is garbled in transit?•Add error detection field in packet?°What if packet is lost?•More elaborate protocols to detect loss?°What if multiple processes/machine?•Queue per process?°Questions such as these lead to more complex protocols and packet formatscs 152 nets.14©DAP & SIK 1995Protocol StacksApplicationPresentationSessionTransportNetworkData linkPhysicalNetworkData linkPhysicalNetworkData linkPhysicalApplicationPresentationSessionTransportNetworkData linkPhysicalSubnetHost A Host BOSI Reference ModelXmit raw bitsFraming, ErrorrecoveryRoutingFlow controlCongestionManage dialogueSynchronizationFrequently usedfunctions (e.g. charconversion)Virtual terminal,File transfer, . . .interfaceIPTCPMediumAccesscs 152 nets.15©DAP & SIK 1995Interconnection Networks°Examples•MPP networks (CM-5): 1000s nodes; Š 25 meters per link•Local Area Networks (Ethernet): 100s nodes; Š 1000 meters•Wide Area Network (ATM): 1000s nodes; Š 5,000,000 meterscs 152 nets.16©DAP & SIK 1995Interconnection Network Issues°Implementation Issues°Performance Measures°Architectural Issues°Practical Issuescs 152 nets.17©DAP & SIK 1995Implementation IssuesInterconnect MPP LAN WANExample CM-5 Ethernet ATMMaximum length 25 m 500 m; copper: 100 mbetween nodes Š5 repeaters optical: 1000 mNumber data lines 4 1 1Clock Rate 40 MHz 10 MHz O 155.5 MHz Shared vs. Switch Switch
View Full Document