《研究生补充资料分布式系统ppt课件.ppt》由会员分享，可在线阅读，更多相关《研究生补充资料分布式系统ppt课件.ppt（51页珍藏版）》请在三一办公上搜索。
1、Distributed System,Instructor:Dr.Ying ZhaoSci.Building 407(Office location)64433221(Phone)64414105(Fax)(Email)Text:Distributed System,Nancy Lynch,Morgan-Kaufmann,1996.,Expectations,Attendance is expected,but not mandatoryIf you miss class,you are responsible for obtaining the lectures from the websi
2、te and/or notes from a classmateYou should come prepared to classRead ahead email me if you are unsure of what to readDownload lectures that are on PowerPoint take notes as well,Problem Sets,Assignments related to the text and lecture materials not formally graded There will be due dates on which th
3、e answers to the assignments will be worked out in class You are strongly encouraged to discuss possible solutions with other class members to bring to class your solutions,Project and Exams,There will be one project worth 25%of your gradeTopics will be assigned by the instructorFormat,content,and d
4、ue dates There will be three exams and a final.Each exam is worth 25%of your grade.You can drop the lowest exam grade.Tentative exam dates:Feb.3rd,Mar.2nd,Apr.6th,Apr.29th(final).,The Rise of Distributed Systems,Computer hardware prices are falling and power increasing.Network connectivity is increa
5、sing.Everyone is connected with fat pipes.It is easy to connect hardware together.,Definition,“A distributed system is a collection of independent computers that appears to its users as a single coherent system.”(Distributed Systems:Principles and Paradigms,written by TanenbaumImportant Points:The m
6、achines are autonomous.Users think they are dealing with a single system.,What is a Distributed System?,Contains independent,autonomous components thatare located at networked computerscommunicate and coordinate through synchronous or asynchronous message passingare linked by software designed to pr
7、oduce an integrated computing facilityappear to the user of the system as a single component,What is Computer Network?,The term computer network means a collection of autonomous computers interconnected by a single technology in this book.The key distinction is that in a distributed system,a collect
8、ion of independent computers appears to its users as a single coherent system.,Why a Distributed System?,Functional distributioncomputers have different functional capabilities yet may need to share resourcesClient/serverData gathering/data processingInherent distribution in application domaincash r
9、egister and inventory systems for supermarket chainscomputer supported collaborative work,Distributed Systems Middleware?,Distributed System Examples,The biggest example is the WWW.The model presented to users to somewhat uniform go to this URL and get a web page.Most users are(or should be)unable t
10、o tell what kind of hardware or software is being used to deliver the web page.For the most part,the client should not make a difference either.,Goals,There are four main goals for distributed systems to make them worth building.Connect users and resourcesMake the system transparentMake the system o
11、penMake the system scalable,Connecting Users and Resources,We want to make it easier for users to access remote resources and to share resources with other users.Examples:printers,files,Web pages,etcA distributed system should also make it easier for users to exchange information.Easier resource and
12、 data exchange could cause security problems a distributed system should deal with this problem.,Transparency,Transparency is the ability of the distributed system to hide that fact that it is actually a distributed system.The more the distributed system looks like a single computer,the more transpa
13、rent it is.There are several different types of transparency that we should consider,Transparency Types,An important goal of a distributed system is to hide the fact that its processesand resources are physically distributed across multiple computers.A distributedsystem that is able to present itsel
14、f to users and applications as if it were only asingle computer system is said to be transparent.For example,resources may be replicated to increase availability or to improve performance by placing a copy close to the place where it is accessed.,Degree of Transparency,We may not be able to hide all
15、 distribution aspects from the user and sometime we shouldnt.Physical limitations can cause situations where it is better for the user to know about the distributed system.We cant expect our morning paper to be in our mailbox at 7am if the paper is coming from a different time zone.,System Openness,
16、Openness describes how well a system offers services according to some standard rules for the distributed system.An open system will have a set of well-defined standards that describe the syntax and semantics for system services.If the system is not open,it will be difficult to expand.,System Scalab
17、ility,A distributed system should be scalable.Scalability can be measured in three different ways:size(add more users and resources)geographically(add users and resources that are far away)administration(easily able to manage if scaled),Scalability Problems,Sizewill present limitations on centralize
18、d resources such as specialized servers,data and algorithmsWe should decentralize our servers,data and algorithms as much as possible.This causes issues with data consistency and data retrieval.,Decentralized Algorithms,Such algorithms should realize:No machine has complete information about the sys
19、tem stateMachine make decisions based only on local informationFailure of one machine does not ruin the algorithmThere is no implicit assumption that a global clock exists,Scaling Techniques,Use only asynchronous communicationDistribute the problem as much as possible to many machines in small chunk
20、sReplicate components across the system(if possible)Cache data where possible(and then solve the cache consistency problems),Why a Distributed System?,Load balancingassign tasks to processors such that the overall system performance is optimizedReplication of processing powerindependent processors w
21、orking on the same taskEconomics collections of microprocessors offer a better price/performance ratio than large mainframes,Distributed System Characteristics,ConcurrencySoftware/hardware components are autonomous component is equivalent to“processor”Components execute concurrent tasks A and B are
22、concurrent if either A can happen before B,or B can happen before A(interleaving semantics)Users work independently&share resourceshardware components(disks,printers)software entities(files,web pages,data objects)Typical problemsabsence of deadlock Program 1 requests resource A and receives it.Progr
23、am 2 requests resource B and receives it.Program 1 requests resource B and is queued up,pending the release of B.Program 2 requests resource A and is queued up,pending the release of A.liveness guarantees,No global clockCoordination is done by message exchangeClose coordination depends on a shared i
24、dea of the time at which the programs actions occurThere are limits to the accuracy with which computers in a network can synchronize their clocksNo global stateGenerally,there is no single process in the distributed system that would have a knowledge of the current global state of the system,Distri
25、buted System Characteristics,Independent failuresNetwork faults can result in the isolation of computers that continue executingA system failure or crash might not be immediately known to other systems,Distributed System Characteristics,Example 1,InternetHeterogeneous network of computers and applic
26、ationsInteraction via message passing for a common means of communicationUsers in different physical locations can make use of services,such as email and ftpThe set of services is open-ended,A Typical Portion of the Internet,Example 2,IntranetsA portion of the Internet that is separately administere
27、dusually proprietaryprovides internal and external servicescan be configured to enforce local security policiesmay use a firewall to prevent unauthorized messages leaving or enteringmay be connected to the internet via a router,A Typical Intranet,Other Examples,Mobile and ubiquitous computingMobile
28、computingcontinuous service is available to the internet,company intranetsex.Cell phone can access simple informationUbiquitous computingcomputing devices will become so pervasive they will not be noticeablewearables,PDAs,digital camerasIssuesdiscovery of resources,eliminating reconfiguration of dev
29、ices from movement,coping with limited connectivity,privacy and security,Portable and handheld devices in a distributed system,Challenges in Distributed System Design,Heterogeneitynetworks(protocols),hardware(data types)operating systems(APIs)programming languages(characters and data structures)impl
30、ementations by different developers(lack of standards)Middlewarecan mask heterogeneityprovides a uniform computational model for use by the programmers of servers and distributed applicationsMobile codetransparency from hard-,software and programming language heterogeneity through virtual machine co
31、ncept,Challenges in Distributed System Design,OpennessThe degree to which new resource-sharing services can be added and be made available for use by a variety of client programsSpecification and documentation of the key software interfaces of the components must be publishedExtension may be at the
32、hardware level by introducing additional computers,Challenges in Distributed System Design,Security of information resourcesEncryption techniques have resolved many challenges with sending sensitive information in a message across a network in a secure mannerContinued challengesDenial of service att
33、acksSecurity of mobile code,Challenges in Distributed System Design,Scalabilitysystem remains effective when there is a significant increase in the number of resources and the number of userscontrolling the cost of physical resourcesadding servers to allow for increased file sharingcontrolling the c
34、ost of performance lossincreases in data storage and accesspreventing software resources from running outtrade-offs between allocating too much vs.too litleavoiding performance bottlenecksalgorithms should be decentralized,Challenges in Distributed System Design,Failure handlingIn distributed system
35、s,some components fail while others continue executingDetected failures can be hidden,made less severe,or toleratedmessages can be retransmitteddata can be written to multiple disks to minimize the chance of corruptionData can be recovered when computation is“rolled back”Redundant components or comp
36、utations tolerate failure,Challenges in Distributed System Design,ConcurrencySeveral clients may attempt to access a shared resource at the same timeebay bidsGenerally multiple requests are handled concurrently rather than sequentiallyAll shared resources must be responsible for ensuring that they o
37、perate correctly in concurrent environment,Challenges in Distributed System Design,Transparencythe system is perceived as a whole rather than as a collection of independent componentsExamples includeAccess transparency*:enables local and remote resources to be accessed using identical operations.Loc
38、ation transparency*:enables resources to be accessed without knowledge of their location.Concurrency transparency:enables several processes to operate concurrently using shared resources without interference between them.Mobility transparency:allows the movement of resources and clients within a sys
39、tem without affecting the operation of users or programs.,Distributed System Architectures,Client-server modelclient processes interact with individual server processes servers processes are in separate host computers clients access the resources the servers manageservers may be clients of other ser
40、versExamplesWeb servers are clients of the DNS serviceDNS translates Internet Domain Names to network addressesSearch engines are both servers(to browser clients)and clients(web crawlers)of other web servers,Client-Server,Multiple Servers,Separate processors interact to provide a service,Peer Proces
41、ses,All processors play a similar role-eliminate servers,Distributed Algorithms,A definition of the steps to be taken by each of the processes of which the system is composed,including the messages transmitted between themTypes of distributed algorithmsInterprocess Communication(IPC)Timing ModelFail
42、ure Model,Distributed Algorithms,Address problems ofresource allocation-deadlock detectioncommunication-global snapshotsconsensus-synchronizationconcurrency control-object implementationHave a high degree ofuncertainty and independence of activitiesunknown#of processes&network topologyindependent in
43、puts at different locationsseveral programs executing at once,starting at different times,operating at different speedsprocessor non-determinismuncertain message ordering&delivery timesprocessor&communication failures,Interprocess Communication,Distributed algorithms run on a collection of processor
44、scommunication methods may be shared memory,point-point or broadcast messages,and RPCMultiple server processes may cooperate with one another to provide a serviceDNS partitioning and replicating its data at multiple servers throughout the InternetPeer processes may cooperate with one another to achi
45、eve a common goala voice conferencing system that distributes streams audio data in a similar manner under real-time constraints,Failure Model,The algorithm might need to tolerate failuresprocessors might stopdegrade gracefullyexhibit Byzantine failuresmay also be failures of communication mechanism
46、s,Timing Model,Different assumptions can be made about the timing of the events in the systemSynchronousprocessor communication and computation are done in lock-stepAsynchronousprocessors run at arbitrary speeds and in arbitrary orderPartially synchronousprocessors have partial information about tim
47、ing,Synchronous Model,Simplest to describe,program,and reason aboutcomponents take steps simultaneously not what actually happens,but used as a foundation for more complexityintermediate comprehension stepimpossibility results care over,Asynchronous Model,Separate components take steps arbitrarilyRe
48、asonably simple to describe-with the exception of liveness guaranteesHarder to accurately programAllows ignorance to timing considerationsMay not provide enough power to solve problems efficiently,Partially Synchronous Model,Some restrictions on the timing of events,but not exactly lock-stepMost realistic modelTrade-offs must be considered when deciding the balance of the efficiency with portability,
