Crossbar switch. Chapter 2: Concepts and Architectures. Traditional Computer Architecture. Computer System Architectures. Flynn Architectures (2)

Transcription

1 Chapter 2: Concepts and Architectures Computer System Architectures Disk(s) CPU I/O Memory Traditional Computer Architecture Flynn, classification of computer systems in terms of instruction and data stream organizations Based on Von-Neumann model (separate processor and memory units 4 machine organizations SISD - Single Instruction, Single Data SIMD - Single Instruction, Multiple Data MISD - Multiple Instruction, Single Data MIMD - Multiple Instruction, Multiple Data Distributed Systems 1 Distributed Systems 2 Flynn Architectures (1) Flynn Architectures (2) CU SISD I PU D Serial Processor I1 D 1 CU 1 PU 1 In D CU n PU n n Multiprocessor and Multicomputer D 1 CU I PU 1 D n Array Processor MIMD SIMD PU n CU control unit PU processor unit I instruction stream D data stream I 1 CU 1 PU 1 I n CU n PU n MISD D real examples - possibly some pipeline architectures Distributed Systems 3 Distributed Systems 4 Processor-Memory Interconnection Network Crossbar switch Access a memory bank through a switch N 2 cross points vs routing switch Distributed Systems 5 Distributed Systems 6 1

2 Multiple stage switch Homogeneous Multicomputer Systems Processor Arrays Grid Hypercube Distributed Systems 7 Distributed Systems 8 Loosely coupled multi -computer systems Closely coupled multi -computer systems M P M P M P network Shared Memory M P M P M P M P I/O Distributed Memory Multi-computer IPC by message passing Typically PC or workstation clusters Physically distributed components Characterized by longer message delays and limited bandwidth Distributed Systems 9 Shared Memory Multiprocessor Processors connected via common bus or fast network Characterized by short message delays, high bandwidth IPC via Shared Memory Distributed Systems 10 Network based Systems Network size: number of nodes N de: n i, 1 i N Distance: d(n i, n j ): number of links between n i and n j Network distance: D = max(d(n i, n j )) Degree: degree(n i ): number of links from/to n i Network topology is an abstract graph to represent the architecture of a network Desired Properties: (1) When network size grows arbitrarily, the network distance increases very slowly: lim D/N = 0 N (2) There exists a constant k, such that degree(n i ) k (3) Routing algorithm is easy to implement and independent of network size (4) When some nodes or links are failed, the network is still connected (with lower performance) (5) Trafic loads are evenly distributed over the network Distributed Systems 11 Distributed Systems 12 2

3 Typical network topologies: Evaluation of network topologies: star ring B-tree star ring B-tree complete regular arbitra ry lim D/N=0 Nfi yes D = 2 D = N-1 D=2log N D = 1 D =v N Do not K degree (n i ) no K = 2 K = 3 K = 4 Do not complete regular arbitrary Routing easy easy easy easy easy hard connectiv ity Even trafic bad bad t good best good Do not no yes no yes yes Do not Distributed Systems 13 Distributed Systems 14 Software Concepts Uniprocessor Operating System System Description Main Goal DOS NOS Middleware Tightly-coupled operating system for multi-processors and homogeneous multicomputers Loosely-coupled operating system for heterogeneous multicomputers (LAN and WAN) Additional layer atop of NOS implementing general-purpose services DOS (Distributed Operating Systems) NOS (Network Operating Systems) Middleware Hide and manage hardware resources Offer local services to remote clients Provide distribution transparency Separating applications from operating system code through a microkernel. Distributed Systems 15 Distributed Systems 16 Distributed Operating System DOS: characteristics (1) Tightly-coupled operating system for multi -processors and homogeneous multi-computers. Strong transparency. Distributed Operating Systems Allows a multiprocessor or multicomputer network resources to be integrated as a single system image Hide and manage hardware and software resources provides transparency support provide heterogeneity support control network in most effective way consists of low level commands + local operating systems + distributed features Inter-process communication (IPC) Distributed Systems 17 Distributed Systems 18 3

4 remote file and device access global addressing and naming trading and naming services synchronization and deadlock avoidance resource allocation and protection global resource sharing deadlock avoidance communication security DOS: characteristics (2) no examples in general use but many research systems: Amoeba, Chorus etc. see Google distributed systems research Network Operating System Loosely-coupled operating system for heterogeneous multi-computers (LAN and WAN). Weak transparency. Distributed Systems 19 Distributed Systems 20 NOS: characteristics Middleware System Network Operating System extension of centralized operating systems offer local services to remote clients each processor has own operating system user owns a machine, but can access others (e.g. rlogin, telnet) no global naming of resources system has little fault tolerance e.g. UNIX, Windows NT, 2000 Additional layer on the top of NOS implementing generalpurpose services. Better transparency. Distributed Systems 21 Distributed Systems 22 Middleware Examples Examples: Sun RPC, CORBA, DCOM, Java RMI(distributed object technology) Built on top of transport layer in the ISO/OSI 7 layer reference model: application (protocol), presentation (semantic), session (dialogue), transport (e.g. TCP or UDP), network (IP, ATM etc), data link (frames, checksum), physical (bits and bytes) Most are implemented over the internet protocols Masks heterogeneity of underlying networks, hardware, operating system and programming languages so provides a uniform programming model with standard services 3 types of middleware: transaction oriented (for distributed database applications) message oriented (for reliable asynchronous communication) remote procedure calls (RPC) the original OO middleware Types of communication Message passing is the general basis of communication in a distributed system: transferring a set of data from a sender to a receiver. Distributed Systems 23 Distributed Systems 24 4

5 Point-to-point Message passing Distributed Shared Memory sender calls send primitive to pass message to sender s buffer communication module transmits the message to the destination destination communication module puts the message to receiver s buffer receiver calls receive primitive to get the message Distributed Systems 25 a) Pages of address space distributed among four machines b) Situation after CPU 1 references page 10 c) Situation if page 10 is read only and replication is used Distributed Systems 26 Comparison between Systems Client-Server Model Item Distributed OS Multiproc. Multicomp. Network OS Middlewarebased OS Degree of transparency Very High High Low High Same OS on all nodes Number of copies of OS 1 N N N Basis for communication Resource management Shared memory Global, central Messages Global, distributed Files Per node Model specific Per node Scalability Moderately Varies Openness Closed Closed Open Open Distributed Systems 27 Distributed Systems 28 Timing interaction between client and server An Example Client and Server (1) header.h Distributed Systems 29 Distributed Systems 30 5

6 An Example Client and Server (2) server.c An Example Client and Server (3) client.c Distributed Systems 31 Distributed Systems 32 Multi-tiered (client-server) Architectures Comparison of different models(1) user terminal computer Interface-level Processing-level Data-level (a) Traditional terminal/machine Distributed Systems 33 Distributed Systems 34 Comparison of different models(2) Internet search engine into three different layers: client Processing-level Interface-level server Processing-level Data-level Client browser Processing Interface (b) two-tiers WEB server Processing Data Data server Processing Data (c) three-tiers Distributed Systems 35 Distributed Systems 36 6

7 Alternative client-server organizations Enterprise(J2EE) Application Model Distributed Systems 37 Distributed Systems 38 A Modern Architecture Advantages of Multi-tiered Architectures An example of horizontal distribution of a Web service. flexibility of selecting different hardware components easy to maintain and manage the system easy to update or re-organize the system easy to enforce different security policies at different levels easy to implement software modules with clearly defined interfaces Distributed Systems 39 Distributed Systems 40 7