Download Fundamentals - Distributed Software Develop | CS 682 and more Study notes Software Engineering in PDF only on Docsity!
Distributed SoftwareDevelopment^ Fundamentals^ Chris BrooksDepartment of Computer ScienceUniversity of San Francisco^ Department of Computer Science — University of San Francisco – p. 1/
??
Outline
-^ Networking overview^ •^ Seven-layer model •^ Intro to Distributed Systems^ •^ Characteristics^ •^ Desirable Properties^ •^ Dealing with Time
Department of Computer Science — University of San Francisco – p. 2/
??
Outline
-^ Networking overview^ •^ Seven-layer model •^ Intro to Distributed Systems^ •^ Characteristics^ •^ Desirable Properties^ •^ Dealing with Time
Department of Computer Science — University of San Francisco
TCP/IP in 30 minutes • Goal: Understand how a network transmits messages atdifferent layers. • How is a network composed? • What really happens when Firefox opens a connection to aweb server? • Note: this will be an overview: for more details, take thenetworking class.^ Department of Computer Science — University of San Francisco – p. 4/
??
Layering
-^ Modern network design takes advantage of the idea of^ layering •^ A particular service or module is constructed as a black box. •^ Users of that service do not need to know its internals, justits interface. •^ This makes it easy to later build new modules (or layers)that use the lower layers. •^ For example, HTTP is built on top of TCP.^ •^ A web browser does not typically need to worry aboutthe implementation of TCP, just that it works. •^ Unlike OO modules, the layers in a networked systemcomprise protocols that span multiple machines.
Department of Computer Science — University of San Francisco – p. 5/
??
The OSI seven-layer model • ISO (a standards body) developed a reference model calledOSI that defines the different layers needed forcommunication, and specifies which should do each job. • The goal is to produce an open protocol that allows forheterogeneous, extensible systems. • A protocol^ is a specification describing the order and formatof messages. • An open protocol is one in which all of this information ispublicly available.^ Department of Computer Science — University of San Francisco
The OSI seven-layer model • Application • Presentation • Session • Transport • Network • Data Link • Physical^ Department of Computer Science — University of San Francisco – p. 7/
??
Message transmission across layers •^ An application (such as a web browser) wants to send amessage to another computer. •^ That application constructs a message and passes it to theapplication layer. •^ The application layer attaches a header to the message andpasses it to the presentation layer. •^ The presentation layer attaches a header and passes it tothe session layer, and so on.
Department of Computer Science — University of San Francisco – p. 8/
??
Message transmission across laye •^ On the other end, the message is received by the physicallayer, who strips off the appropriate header and passes themessage up to the data link layer. •^ This continues until the message reaches the applicationlayer of the receiving machine. •^ High-level layers don’t need to worry about lower-levellayers. •^ Lower-level layers treat everything from higher layers as datato be sent.
Department of Computer Science — University of San Francisco
Layers and packets • Each layer constructs a packet containing a portion of thedata to be transmitted. • This packet has a data section, and a header. • The header contains origin and destination information,checksums, sequence numbers, and other identifyinginformation. • When a message is sent by TCP, a packet is constructedand passed down to the IP layer. • This entire packet then becomes the data portion of the IPpacket, which is passed down to the network layer, and soon. • On the other end, the lowest layer removes the header andchecks the data integrity, then passes the data portion up tothe next layer.^ Department of Computer Science — University of San Francisco – p. 10/
??
Physical Layer
-^ This is the lowest-level layer, responsible for transmitting 0sand 1s. •^ Governs transmission rates, full or half-duplex, etc. •^ A modem works at the physical layer. •^ Lots of interesting problems at this level that we won’t getinto ...
Department of Computer Science — University of San Francisco – p. 11/
??
Data Link Layer • The data link layer provides error handling for the physicallayer. • Individual bits are grouped together into frames. • A checksum is then computed to detect transmission errors. • The data link layer can then request a retransmission of anerror is detected. • Messages are numbered; receiver can requestre-transmission of any message in a sequence. • Each frame is a separate, distinct message. • The Data link layer provides error-free transmission toupper-level layers.^ Department of Computer Science — University of San Francisco –
An example: HTTP • HTTP is the protocol that drives the Web. • A side note/axe to grind: WWW != Internet!! • It is a stateless protocol that uses TCP as its underlyingprotocol. • The client sends a request, which is processed by theserver. • The server sends a reply, and the exchange is ended.^ Department of Computer Science — University of San Francisco – p. 19/
??
HTTP requests • HTTP has a very simple message format. GET /~brooks/index.html HTTP/1.1Host: www.cs.usfca.eduConnection: closeUser-agent: Mozilla/4.0Accept-language: en • You can try this out for yourself with telnet ...^ Department of Computer Science — University of San Francisco – p. 20/
??
HTTP
-^ There are lots of wrinkles and extensions to HTTP^ •^ Cookies to help save state^ •^ CGI, SOAP to pass data and execute code as the resultof an HTTP request.^ •^ Web caching to store data closer to clients. •^ These are all possible because HTTP is an open protocol. •^ This is also what makes it possible for different companiesto write web browsers and web servers that seamlessly worktogether.
Department of Computer Science — University of San Francisco –
Summary
-^ The modern networking stack can be conceptually brokeninto a set of layers. •^ Each layer has a specific, well-defined function.^ •^ Acts as a black box •^ Higher-level layers build on the functionality of lower-levellayers. •^ We’ll be primarily concerned with the Transport andApplication layers.
Department of Computer Science — University of San Francisco – p. 22/
??
Outline
-^ Networking overview^ •^ Seven-layer model •^ Intro to Distributed Systems^ •^ Characteristics^ •^ Desirable Properties^ •^ Dealing with Time
Department of Computer Science — University of San Francisco – p. 23/
??
What is a Distributed System? • What is a distributed system? •^ (Couloris) “A distributed system is one in whichhardware or software components communicate orcoordinate their actions only by passing messages.” •^ (Tanenbaum) “A distributed system is a collection ofindependent computers that appear to the users of thesystem as a single computer.” •^ (Lamport) “You know you have one when the crash of acomputer you’ve never heard of stops you from gettingany work done.” • All of these get at different aspects of the issue ...
Department of Computer Science — University of San Francisco –
Advantages of a distributed system^ •^ Can share expensive resources or data^ •^ Economics^ •^ A collection of PCs can provide betterprice/performance than a single mainframe.^ •^ Speed^ •^ A distributed system will often have more computingpower than a single mainframe.^ •^ Inherent distribution^ •^ Often, your data/users/resources are geographicallydistributed
Department of Computer Science — University of San Francisco – p. 25/
??
Advantages of a distributed system •^ Reliability^ •^ If one node fails, the rest of the system can continue •^ Incremental growth^ •^ Components can be added or replaced in smallincrements.
Department of Computer Science — University of San Francisco – p. 26/
??
Disadvantages of distributed syste^ •^ Software design is much more complicated.^ •^ Lack of appropriate tools/languages^ •^ Disagreement on principles: how much should usersknow about the system? How much the system handleon a user’s behalf?^ •^ Potential network saturation^ •^ Privacy and security issues^ •^ Allowing resources to be shared can lead to data leakage^ •^ Extra sysadmin work
Department of Computer Science — University of San Francisco –
Design Issues
-^ Transparency •^ Flexibility •^ Dependability •^ Performance •^ Scalability
Department of Computer Science — University of San Francisco – p. 28/
??
Transparency
-^ The goal of transparency is a
single-system image
-^ From the user’s POV, it looks like a single machine. • Types of transparency: •^ Location transparency - Users cannot tell where theirresources are actually located. •^ Migration transparency - Resources can move withoutchanging their names. •^ Replication transparency - the number of copies of aresource is hidden from users. •^ Concurrency transparency - Users can share resourceswithout being aware of other users. •^ Parallelism transparency - A task can be run on multiplemachines without the user being aware of it.
Department of Computer Science — University of San Francisco – p. 29/
??
Transparency
-^ Is transparency always a good thing? What is the downside?
Department of Computer Science — University of San Francisco –
Communication paradigms • Asynchronous: there is no bound on message delay • Synchronous: • Known upper bound^
b^ on message delay
-^ Every process^ p^
has a local clock^ C
which drifts at ap rate of^ r >^0 and^ ∀
′p and ∀t > t: C(t)−C− 1 p (1 + r)≤ ′(t)p≤^ (1 +^ r)′ (^) t−t
-^ In English, clock drift has an upper and lower bound. •^ Also, bounds on the amount of time needed for aprocess to execute a single step. • Synchronous communication allows you to implementapproximately synchronized clocks, even in the presence offailure.
Department of Computer Science — University of San Francisco – p. 37/
??
Dealing with time • One of the fundamental problems in distributed systems isdealing correctly with time. • Not only when things happened, but what order thingshappened in. • We would like for all processes to see
relevant^ chages in the same order.^ •^ Example: updating a replicated database. • Depending on the communication model, this may be quitedificult. • Insight: often, it doesn’t matter exactly what time anoperation happens, but what order events occur in. • (exception: hard real-time systems)
Department of Computer Science — University of San Francisco – p. 38/
??
Global time servers • NTP is an Internet Protocol that allows your machine tosynchronize its clock with a remote source, thereby keepingit accurate. • Is that all we need to do?^ Department of Computer Science — University of San Francisco –
Global time servers • NTP is an Internet Protocol that allows your machine tosynchronize its clock with a remote source, thereby keepingit accurate. • Is that all we need to do? • Maybe. Maybe not. • What if we don’t have an Internet connection, or NTPis blocked by our firewall? • Can we guarantee that all users use the same remotetime server? • How often should they update? • What if users don’t do this?^ Department of Computer Science — University of San Francisco – p. 40/
??
Logical time
-^ The algorithms we’ll look at in this class will not need todepend on the^ absolute
time that something happens.
-^ Instead, we’ll be interested in the
logical^ time, or^ causal order^ in which events occur. • As long as all processes agree on the order in which a set ofevents that influence each other occurs, we’re OK. • We’ll spend time next week looking at this problem.
Department of Computer Science — University of San Francisco – p. 41/
??
Summary
-^ There are lots of desirable properties and design issues fordistributed systems.^ •^ Performance, scalability, reliability, flexibility,transparency^ •^ Often, we must sacrifice one for another^ •^ Some (e.g. Parallel transparency) are not possible withtoday’s technology. •^ Communication can be either
synchronous^ or^ asynchronous
-^ Time is a very sticky problem to deal with in distributedsystems. •^ Characterizing types of failure will help us identify what ouralgorithms and systems can and cannot stand up to.
Department of Computer Science — University of San Francisco –
