Chapter 1: Characterization of Distributed Systems Introduction Examples of distributed systems Resource sharing and the eb Challenges Summary
!bi"uitous net or#s Internet $obile phone net or#s Corporation net or#s %actory net or#s Campus net or#s &ome net or#s ''
Distributed System Definition ( distributed system is one in hich hard are or soft are components located at net or#ed computers communicate and coordinate their actions only by passing messages)
Characteristics of Distributed System Concurrency
concurrent programs execution * share resource
+o global cloc#
programs coordinate actions by exchanging messages
Independent failures
hen some systems fail, others may not #no
Share resources
It characterizes the range of the things that can usefully be shared in a net or#ed computer
It extends from hard are components to soft are-defined entities) It includes the stream of .ideo frames and the audio connection)
Chapter 1: Characterization of Distributed Systems Introduction Examples of distributed systems Resource sharing and the eb Challenges Summary
%amiliar and idely used computer net or#s /he Internet Intranet $obile computing
Internet
It is a .ery large distributed system that allo s users throughout the orld to ma#e use of its [Link])
Internet protocols is a ma0or technical [Link])
/he Internet
intranet ISP
backbone
satellite link desktop computer: server: network link:
Intranet 1hat is Intranet2
* ( portion of the Internet that is separately administered and has a boundary that can be configured to enforce local security policies * Composed of [Link] 3(+s lin#ed by bac#bone connections * 4e connected to the Internet .ia a router
( typical Intranet
email server print and other servers Local area network
Des#top computers
Web server
email server File server print other servers the rest of the Internet router/firewall
/hree main issues in the design of components for the use in intranet %ile [Link] %ire all /he cost of soft are installation and support
$obile and ubi"uitous computing $obile [Link]
* 3aptop computers * &andheld [Link]
5D(, mobile phone, pager, .ideo camera, digital camera
* 1earable [Link]
e)g) smart atches, digital glasses
* +et or# appliances
e)g) ashing machines, hi-fi systems, cars and refrigerators
$obile and ubi"uitous computing ' continued $obile computing 6nomadic computing7
* (ccess resources hile on the mo.e or in an unusual [Link] * Location-aware computing: utilize resources that are [Link] nearby
!bi"uitous computing [Link].e computing7
* /he harnessing of many small, cheap computational [Link]
5ortable and handheld [Link] in a distributed system
Internet
ost intranet
Wireless L!"
W!P #atew a$
ome intranet
Printer Camera
Mobile phone Laptop
ost site
Issues in the design of components for the use in $obile and ubi"uitous computing [Link] of resources Eliminating the need for users to reconfigure their mobile [Link] /o cope ith limited [Link] as they [Link] [Link] [Link] and other security guarantees
Chapter 1: Characterization of Distributed Systems Introduction Examples of distributed systems Resource sharing and the eb Challenges Summary
Resource sharing Is the primary [Link] of distributed computing Resources types
* &ard are, e)g) printer, scanner, camera * Data, e)g) file, database, eb page * $ore specific functionality, e)g) search engine
Some definitions Service
* manage a collection of related resources and present their functionalities to users and applications
Server
* a process on net or#ed computer that accepts re"uests from processes on other computers to perform a service and responds appropriately
Client
* the re"uesting process
Remote invocation
* ( complete interaction bet een client and server, from the point hen the client sends its re"uest to hen it [Link] the ser.er8s response
Case study: the 1orld 1ide 1eb $[Link] of 111
* Documents sharing bet een physicists of CER+
1eb is an open system: it can be extended
and implemented in ne ays ithout disturbing its existing functionality)
Its operation is based on communication standards and document standards * Respect to the types of 9resource8 that can be published and shared on it)
/hree main components of the 1eb &yper/ext $ar#up 3anguage
* ( language for specifying the contents and layout of pages as they are displayed by eb bro sers
!niform Resource 3ocators
Identify documents and other resources stored as part of the 1eb
( [Link] architecture ith &//5
4y ith bro sers and other clients fetch documents and other resources from eb [Link]
&/$3
<IMG SRC = [Link] <P> Welcome to Earth! Visitors may also be interested in taking a look at the <A HREF = [Link] <P> (etcetera)
&/$3 text is stored in a file of a eb [Link]) ( bro ser [Link] the contents of this file from a eb [Link]) -/he bro ser interprets the &/$3 text -/he [Link] can infer the content type from the filename extension)
!R3
S c h e m e : sc h e m e -sp e c ific -lo c atio n e .g: m ailto :jo e @ an IS P .n e t ftp ://ftp .d o w n lo ad It.c o m /so ftw ar e /aP r o g.e x e h ttp ://n e t.p k u .e d u .c n / .
&//5 !R3s are the most idely used (n &//5 !R3 has t o main 0obs to do: - /o identify hich eb [Link] maintains the resource - /o identify hich of the resources at that [Link]
1eb [Link] and eb bro sers
[Link]
[Link] Web servers [Link] [Link] File system o [Link] Protocols [Link] Internet [Link] Browsers
[Link]
&//5 !R3s
http:::[Link];:port<::path+ame=[Link]<;2arguments< e)g) http::: )cd#>)net: http::: ) >c)org:5rotocols:([Link])html http:::e)p#u)cn:cgi-bin:allsearch2 ord?distributed@system
--------------------------------------------------------------------------------------
[Link] D+S name )cd#>)net ) >c)org e)p#u)edu)cn
5athname on [Link] 6default7 5rotocols:([Link])html cgi-bin:allsearch
(rguments 6none7 6none7 ord?distributed@system
--------------------------------------------------------------------------------------
5ublish a resource remains un ieldy
&//5 Defines the ays in hich bro sers and any other types of client interact ith eb [Link] 6R%CAB1B7 $ain features
* Re"uest-replay interaction * Content types) /he strings that denote the type of content are called $I$E 6R%CACDE,ACDB7 * =ne resource per re"uest) &//5 .ersion 1)C * Simple access control
$ore [Link] and dynamic pages Dynamic content
* Common Fate ay Interface: a program that eb [Link] run to generate content for their clients
Do nloaded code
* [Link] * (pplet
Discussion of 1eb
Dangling: a resource is deleted or [Link], but lin#s to it may still remain %ind information easily: e)g) Resource Description %rame or# hich standardize the format of metadata about eb resources Exchange information easily: e)g) H$3 * a self describing language Scalability: hea.y load on popular eb [Link] $ore applets or many images in pages increase in the do nload time
Chapter 1: Characterization of Distributed Systems Introduction Examples of distributed systems Resource sharing and the eb Challenges Summary
&eterogeneity +et or#s
* Ethernet, to#en ring, etc
Computer hard are
* big endian : little endian
=perating systems
* different (5I of !nix and 1indo s
5rogramming languages
* different representations for data structures
Implementations from different [Link]
* no application standards
&eterogeneity' continued $iddle are
* applies to a soft are layer that [Link] a programming abstraction as ell as mas#ing the heterogeneity of the underlying net or#s, hard are, =Ss and programming languages
$obile code
* is used to refer to code that can be sent from one computer to another and run at the destination
=penness =penness of a computer system -- is the characteristic that determines hether the system can
be extended and re-implemented in .arious ay) e)g) !nix
=penness of distributed systems -- is determined by the degree to hich ne resource
sharing [Link] can be added and be made [Link] for use by a .ariety of client programs. e.g. Web
&o to deal ith openness2 -- #ey interfaces are published, e)g) R%C
Security Confidentiality
protection against disclosure to unauthorized [Link], e)g) (C3 in !nix %ile System protection against alteration or corruption, e)g) chec#sum protection against interference ith the means to access the resources, e)g) Denial of [Link]
Integrity
(.ailability
Scalability
( system is described as scalable
* if ill remain effecti.e hen there is a significant increase in the number of resources and the number of users
( scalable example system: the Internet design challenges
% % % % Date 1II>, Guly 1IIE, Guly 1IIJ, Guly 1III, Guly /he cost of physical resources, e)g), [Link] support users at most =6n7 /he performance loss, e)g), D+S no orse than =6logn7 [Link] soft are resources running out, e)g), I5 address (.oid performance bottlenec#s, e)g), partitioning name table of D+S, cache and replication Computers 1,JJB,CCC B,BDA,CCC 1I,EDC,CCC EB,A1K,CCC Web servers 1>C A>,ECC 1,AC>,CIB B,EIK,BIJ Percentage C)CCK C)D B 1A
%ailure handling
Detecting
* e)g) chec#sum for corrupted data * Sometimes impossible so suspect, e)g) a remote crashed [Link] in the Internet
Masking
* e)g) Retransmit message, standby [Link]
Tolerating
* e)g) a eb bro ser cannot contact a eb [Link]
Recovery
* e)g) Roll bac#
Redundancy
* e)g) I5 route, replicated name table of D+S
Concurrency Correctness
* ensure the operations on shared resource correct in a concurrent [Link] e)g) records bids for an auction
5erformance
* Ensure the high performance of concurrent operations
/ransparency Access transparency
* using identical operations to access local and remote resources, e)g) a graphical user interface ith folders resources to be accessed ithout #no ledge of their location, e)g) !R3 [Link] processed operate concurrently using shared resources ithout interference ith bet een them multiple instances of resources to be used to increase reliability and performance ithout #no ledge of the replicas by users or application programmers, e)g) realcourse6http ://re a lc ou rse .grids.c n/ 7
Location transparency
*
Concurrency transparency
*
Replication transparency
*
/ransparency ' continued
Failure transparency
* users and applications to complete their tas#s despite the failure of hard are and soft are components, e)g), email [Link] of resources and clients ithin a system ithout affecting the operation of users and programs, e)g), mobile phone allo s the system to be reconfigured to impro.e performance as loads .ary allo s the system and applications to expand in scale ithout change to the system structure or the application algorithms
Mobility transparency
*
Performance transparency
*
Scaling transparency
*
Chapter 1: Characterization of Distributed Systems Introduction Examples of distributed systems Resource sharing and the eb Challenges Summary
Summary
Distributed systems are [Link].e Resource sharing is the primary [Link] for constructing distributed systems Characterization of Distributed System
* * * * * * * * * * Concurrency +o global cloc# Independent failures &eterogeneity =penness Security Scalability %ailure handling Concurrency /ransparency
Challenges to construct distributed system