CS 582 / CMPE 481 Distributed Systems Communications

Class Overview What is a network? Protocols Layered architecture Reference models Protocols for distributed systems

Networks Definitions –A set of Machines (computers, phones, routers, switches, gateways, etc.) connected by communications links –A communication subsystem Why Networks? –Sharing resources –Increasing proximity between people/machines

Requirements Performance –Transmission delay is arbitrary but finite –Transmission time Latency + message size / data transfer rate Reliability –Hardware medium does not support replication –Error recovery Message lost Message duplication Message out-of-order Message corruption

History of Networking Initially circuit switching (PSTN) 1960’s: Packet Switching –ARPAnet (1969) 1970’s: LAN –Ethernet 1980’s: Workstations & PCs –Proliferation of LANs & WANs 1990’s: WWW –Explosion of Internet nodes –Mobile networks

Network Basics Topologies –star, bus, mesh, ring, etc. Switching –circuit, packet, and message switching Transmission technology –point-to-point vs. broadcasting Types of networks –LAN, MAN, and WAN Internetworking –Router –Gateway –bridge

Protocols A protocol specifies –rules and formats used for interactions between two parties specification of the sequences of messages specification of the format of the data in the message Requirements –specifications must be represented in finite states –implementation independent

Layered Architecture Motivation –reducing complexity by modularizing tasks vertically –encapsulation by a well-defined service interface to the upper layer layer N provides a service to layer N+1 layer N extends the service of layer N-1 –independence of each layer allows various implementations across layers

ISO – OSI Reference Model

Services in Layered Protocols Data Transport Service types –Connection-oriented vs. connectionless –Virtual circuit vs. datagram –Depends on QoS requirements of layer above Packet assembly –message at layer N+1 is fragmented into multiple PDUs at layer N (MTU at layer N+1 > MTU at layer N) –layer N at peer: responsible for assembly Addressing –identification of the peer at layer N –conversion into layer N-1 address

Protocol Design Issues Quality of service –reliable vs. unreliable transmission –messages vs. stream –error-free vs. error-durable Error recovery –data errors causes: network error, malicious forgery, etc. detection and correction: CRC –transmission errors causes: network failure, overflow, receiver busy, etc. detection and correction: sequence number and retransmission Flow control –stop and wait; sliding window

Layers –Low-level layers –Transport layer –Application layer –Middleware layer

Low Level Layers Physical layer: contains the specification and implementation of bits, and their transmission between sender and receiver Data link layer: prescribes the transmission of a series of bits into a frame to allow for error and flow control Network layer: describes how packets in a network of computers are to be routed. Observation: for many distributed systems, the lowest level interface is that of the network layer.

Transport Layer The transport layer provides the actual communication facilities for most distributed systems. Many application protocols are directly implemented on top of transport protocols, doing a lot of application-independent work. Standard Internet protocols: –TCP: connection-oriented, reliable, stream-oriented communication –UDP: unreliable (best-effort) datagram communication

Multicast protocols Definition –a means of one-to-many communication unicast (one-to-one) multicast (one-to-some) broadcast(one-to-all)

Overview Two main tasks in multicast –group management how can group membership be captured on a leaf subnet eg: IGMP –multicast routing how can packets be delivered from the source to the group members in an efficient manner ensures the generation of only as many copies of a packet as are necessary to reach the group members algorithms are broadly categorized –shortest path tree –minimum cost tree –constrained tree