Operating Systems COT 4600 – Fall 2009Lecture 3NamesNames and fundamental abstractionsComputers a distinct species of complex systemsAnalog, digital, and hybrid systemsComputers are controlled by softwareExponential growth of computersCoping with complexity of computer systemsIteration – design principlesCurbing complexityCritical elements of information revolution!The relation between homo sapiens and the computersResource sharing and complexityOperating Systems COT 4600 –Fall 2009Dan C. MarinescuOffice: HEC 439 BOffice hours: W, F 3:00-4:00 PMLecture 12z Last timez Computer Systemsz Today:z Namesz Computer Systems versus Other Systemsz Coping with Computer System Complexityz Next timez The projectz Homework 1 due Thursday, September 3, 2009Lecture 3Lecture 13Namesz Modularity along with abstraction, layering, and hierarchy allow a designer to cope with complexity;z Names and addresses Î provide the means to connect modules. z A system Î a bunch of resources, glued together with namesz Naming allows the designer to:z Delay the implementation of some modules; use dummy onesz Replace an implementation with another one.z BindingÎ choosing an implementation for a modulez Delayed binding; use a place holder.Lecture 14Names and fundamental abstractionsz The fundamental abstractions1. Storage Î mem, disk, data struct, File Systems, disk arrays2. Interpreters Î cpu, programming language e.g. java VM3. CommunicationÎ wire, Ethernetrely on names.z Naming:z Flatz HierarchicalLecture 15Computers a distinct species of complex systemsz The complexity of computer systems not limited by the laws of physics Î distant bounds on compositionz Digital systems are noise-free.z The hardware is controlled by softwarez The rate of change unprecedentedz The cost of digital hardware has dropped in average 30% per year for the past 35 yearsLecture 16Analog, digital, and hybrid systemsz Analog systems:z the noise from individual components accumulate and z the number of components is limitedz Digital systems:z are noise-freez the number of components is not limitedz regeneration Î restoration of digital signal levelsz static disciplineÎ the range of the analog values a device accepts for each input digital value should be wider than the range of analog output valuesz digital components could fail but big mistakes are easier to detect than small ones!!z Hybrid systems Î e.g., quantum computers and quantum communication systemsLecture 17Computers are controlled by softwarez Composition of hardware limited by laws of physics.z Composition of software is not physically constrained; z software packages of 107lines of code existz Abstractions hide the implementation beneath module interfaces and allow the z creation of complex softwarez modification of the modulesz Abstractions can be leaky. Example, representation of integers, floating point numbers.Lecture 18Exponential growth of computersz Unprecedented: z when a system is ready to be released it may already be obsolete. z when one of the parameters of a system changes by a factor of z 2 Îother components must be drastically altered due to the incommensurate scaling. z 10 Î the systems must be redesigned; E.g.; balance CPU, memory, and I/O bandwidth;z does not give pause to developers z to learn lessons from existing systemsz find and correct all errorsz negatively affects “human engineering” Î ability to build reliable and user-friendly systemsz the legal and social frameworks are not readyLecture 19Coping with complexity of computer systemsz Modularity, abstraction, layering, and hierarchy are necessary but not sufficient.z An additional technique Î iterationz Iterationz Design increasingly more complex functionality in the systemz Test the system at each stage of the iteration to convince yourself that the design is soundz Easier to make changes during the design processLecture 110Iteration – design principlesz Make it easy to changez the simplest version must accommodate all changes required by successive versionsz do not deviate from the original design rationalez think carefully about modularity Î it is very hard to change it.z Take small steps; rebuild the system every day, to discover design flaws and errors. Ask others to test it.z Don’t rush to implementation. Think hard before starting to program.z Use feedback judiciouslyÎz use alpha and beta versionsz do not be overconfident from an early successz Study failures Î understand that complex systems fail for complex reasons.Lecture 111Curbing complexityz In absence of physical laws curb the complexity by good judgment. Easier said than done because:z tempted to add new features than in the previous generationz competitors have already incorporated the new features z the features seem easy to implementz the technology has improvedz human behavior: arrogance, pride, overconfidence…Lecture 112Critical elements of information revolution!Lecture 113Lecture 114The relation between homo sapiens and the computersz The feelings of the homo sapiens:z Hatez Frustrationz Lack of understandingz The Operating System z A program to “domesticate” the computer.z Transforms a “bare machine” into a “user machine”z Controls and facilitates access to computing resources; optimizes the use of resources.z The relation went through several stages:z Many-to-onez One-to-onez Many-to-manyz Peer-to-peerLecture 115Resource sharing and complexityz A main function of the OS is resource sharing.z Sharing computer resources went through several stages with different levels of complexity:z Many-to-onez One-to-onez Many-to-manyz Peer-to-peerLecture 116Lecture 117HS – Homo SapiensPC- Personal Computer