1Complexity revisited:learning from failuresLec 26 --- Last one!5/16/07Credit: Jerry Saltzer6.033 in one slide• Client/server• RPC• File abstraction• Virtual memory• Threads• Coordination• Protocol layering• Routing protocols• Reliable packet delivery• Names• Replication protocols• Transactions• Verify/Sign• Encrypt/Decrypt• ACL and capabilities• Speaks forCase studies of successful systems: LISP, UNIX, X Windows, MapReduce, Ethernet, Internet, WWW, RAID, DNS, ….Principles:End-to-end argument, Open Design, …Today:Why do systems fail anyway?• Complexity in computer systems has no hard edge• Learning from failures: common problems• Fighting back: avoiding the problems• Admonition & 6.033 theme songToo many objectives• Ease of use• Availability• Scalability• Flexibility• Mobility• Security• Networked• Maintainability• Performance• Durable• ….Lack systematic methodsMany objectives+Few Methods+High d(technology)/dt=Very high risk of failureQuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.The tarpit[Brooks, Mythical Man Month]Complexity: no hard edge• It just gets worse, worse, and worse …Increasing functionSubjectivecomplexity2Learn from failure“The concept of failure is central to design process, and it is by thinking in terms of obviating failure that successful designs are achieved…”[Petroski]Keep digging principle• Complex systems systems fail for complex reasons– Find the cause …– Find a second cause …– Keep looking …– Find the mind-set. [Petroski, Design Paradigms]QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.Try 1: Meidum (52° angle)QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.Try 2: Dashur/Bent (52° to 43.5° angle)QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.Try 3: Red pyramid (right angle: 43°)Pharaoh Sneferu’s Pyramid projectUnited Airlines/Univac• Automated reservations, ticketing, flight scheduling, fuel delivery, kitchens, and general administration• Started 1966, target 1968, scrapped 1970, spend $50M• Second-system effect (First: SABRE)(Burroughs/TWA repeat)CONFIRM• Hilton, Marriott, Budget, American Airlines• Hotel reservations linked with airline and car rental• Started 1988, scrapped 1992, $125M• Second system• Dull tools (machine language)• Bad-news diode[Communications of the ACM 1994]IBM Workplace OS for PPC• Mach 3.0 + binary compatability with AIX + DOS, MacOS, OS/400 + new clock mgmt + new RPC + new I/O + new CPU• Started in 1991, scrapped 1996 ($2B)• 400 staff on kernel, 1500 elsewhere• “Sheer complexity of class structure proved to be overwhelming”• Inflexibility of frozen class structure• Big-endian/Little-endian not solved[Fleish HotOS 1997]3Advanced Automation System• US Federal Aviation Administration• Replaces 1972 Air Route Traffic Control System• Started 1982, scrapped 1994 ($6B)• All-or-nothing• Changing specifications• Grandiose expectations• Contract monitors viewed contractors as adversaries• Congressional meddlingLondon Ambulance Service• Ambulance dispatching• Started 1991, scrapped in1992 (20 lives lost in 2 days, 2.5M)• Unrealistic schedule (5 months)• Overambitious objectives• Unidentifiable project manager• Low bidder had no experience• No testing/overlap with old system• Users not consulted during design[Report of the Inquiry Into The London Ambulance Service 1993]More, too many to list …• Portland, Oregan, Water Bureau, 30M, 2002• Washington D.C., Payroll system, 34M 2002• Southwick air traffic control system $1.6B 2002• Sobey’s grocery inventory, 50M, 2002• King’s County financial mgmt system, 38M, 2000)• Australian submarine control system, 100M, 1999• California lottery system, 52M• Hamburg police computer system, 70M, 1998• Kuala Lumpur total airport management system, $200M, 1998• UK Dept. of Employment tracking, $72M, 1994• Bank of America Masternet accounting system, $83M, 1988,• FBI virtual case, 2004.• FBI Sentinel case management software, 2006.Recurring problems• Excessive generality and ambition• Bad ideas get included• Second-system effect• Mythical Man Month• Wrong modularity • Bad-news diode• Incommensurate scalingFighting back: control novelty• Source of excessive novelty:– Second-system effect– Technology is better– Idea worked in isolation– Marketing pressure•Somenovelty is necessary; the difficult part is saying No.• Don’t be afraid to re-use existing components– Don’t reinvent the wheel– Even if it takes some massagingFighting back: adopt sweeping simplifications• Processor, Memory, Communication• Dedicated servers• N-level memories• Best-effort network• Delegate administration• Fail-fast, pair-and-compare• Don’t overwrite• Transactions• Sign and encrypt4Fighting back:design for iteration, iterate the design• Something simple working soon– Find out what the real problems are• One new problem at a time• Use iteration-friendly design– E.g., Failure/attack models“Every successful complex system is found to have evolved from a successful simple system”Fighting back: find bad ideas fast• Question requirements– “And ferry itself across the Atlantic” [LHX light attack helicoper]• Try ideas out, but don’t hesitate to scrap• Understand the design loopRequires strong, knowledgeable managementThe design loop• Find flaws fast!InitialdesignDraft design coding testing deployedmonthsminhoursdaysweeksFighting back: find flaws fast• Plan, plan, plan (CHIPS, Intel processors)• Simulate, simulate, simulate – Boeing 777 and F-16• Design reviews, coding reviews, regression tests, daily/hourly builds, performance measurements• Design the feedback system:– Alpha and beta tests– Incentives, not penalties, for reporting errorsFighting back:conceptual integrity• One mind controls the design– Macintosh– Visicalc spreadsheet– UNIX– Linux• Good esthetics yields more successful systems– Parsimonious, Orthogonal, Elegant, Readable, …• Few top designers can be more productive than a larger group of average designers.Summary• Principles that help avoiding failure– Limit novelty– Adopt sweeping simplifications– Get something simple working soon– Iteratively add capability– Give incentives for
View Full Document
Unlocking...