All exponentials must end K 2 573725771157630576165775457740577255771857693576165774457347576305779857693577445775457630576265734757615576305764057725577405763057347D57725577255774057630582035774457347 3 There will be electronics after it Mathforumo ID: 18021
Download Pdf The PPT/PDF document "Distribution Statement A Approved for Pu..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) Robert Colwell Director, Microsystems Technology Office DARPA The Chip Design Game at the End of aooreâs Law Hot Chips, August 2013 Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 2 Mooreâs Law Best tech exponential improvement trend ever ! From this⦠â¦to this. 1955 2010 © Texas Instruments © Intel Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 3 How Did We Do That? ⢠Separation of concerns â ISA, microarchitecture, functional blocks, circuits, placement, routing, layout, silicon, packaging â All could be done by specialists ⢠Tools â Checked & enabled this separation of concerns â CMOS was just plain beautiful technology ⢠Simple, small, reliable, fast, low power, high yieldÍvery hard to beat! © Intel Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 4 Electronics Will Go On⦠1. All exponentials must end â hr they will âeat the universeâ 2. Electronics existed before aooreâs Law 3. There will be electronics after it Íbut things are changing rapidly! © Mathforum.org Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 5 Canonical âMooreâs Lawâ Chip Designerâs Algorithm Given constraints { Si area, schedule, team, tools, power } How can I best design a chip to sell huge unit volumes at high yields and high profits? People need a reason to part with their money. Ltâs the chip designerâs job to give them that reason. Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 6 So, is Death of Mooreâs Law a Big Deal? ⢠aooreâs Law silicon improvements are âfor freeâ to designers ⢠Chip architectâs job was to remove obstacles from aooreâs Law silicon bounty â System perf = f(CPU clock, CPU uArch, sys) â From 1980 to 2010, clocks are 3,500x faster â What did Arch/uArch achieve beyond that? Maybe 50x? ⢠Can we continue to crank out successful new chips with only 10% wins per new gen? â Remaining generations of silicon down to 7nm are not nearly as good as previous gens â weâre already power - constrainedÍthat wonât improve YES. Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 7 What I Do NOT Agree With âWe may be getting close to the end of aoore's law as far as single chip performance but I think it will be a long time before we see the end of aooreâs law. While CPU's and GPU's may be getting close to their physical limits, there are other ways computers can still be made fasterÍ Ultimately, L think that aooreâs Law should never stop. Computer builders will just find other methods to make computers faster and more efficient . If there is a demand I think we will find a way to make it work.â http://www.pcworld.com/article/2032913/the - end - of - moores - law - is - on - the - horizon - says - amd.html Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 8 Partial Truths âThe end of aooreâs Law is always 10 years away,â said Lntel's lead researcher aark .ohr. âAnd yes, itâs still 10 years away.â Would you have liked us to stop back when telephones had to plug into walls? Didnât think so. Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 9 All Post - Mooreâs - Law Roads Are Not Blocked a) 3D stacking & improved packaging b) Better cooling, longer battery life c) Software, uArch, Arch, I/O, memory d) New apps e) Resilience (tough sell) f) Alternative switch technologies g) DARPA PERFECT & UPSIDE: beyond convention h) Better system features (sensors, wireless, displays, human interface)Íwhatever works i) aarketing (donât laugh, you will see this) © Intel Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 10 But It Will Be Anything But Easy CMOS has given us the illusion of binary digital crispness for several decades, but the gameâs up â Transmission lines, high - frequency design â Ground bounce, ground loops, power supply design, VRegs â Electromagnetics (inductances) â Crosstalk â Metastability â Thermals â Resilience and software handlers â EFI and RFI © Artofillusion.org Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 11 Computer Design: Whatâs Next No more helpful separation of concerns â Resilience, thermals, battery life, complexity, validation, performance, schedule, riskÍ â¢ Have to juggle them all at once , pre - silicon (but remember i432!) ⢠Meanwhile RTL correctness is just as hard as ever â Specialized HW to go after 2 - 3 orders of magnitude boost in efficiency/perf may again become viable ⢠Dedicated blocks, approx computing, analog © OBYX.org Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 12 Future Designers Must Master Chips AND âNeighboring Technologiesâ Must absorb more of overall value proposition â Lesson of âLntel LnsideâÍ get buyer to relate perceived value to your part of final product â Lessons from Apple: iTunes, white earbuds â This requires expertise beyond chip/CPU design 1. Communications 2. Biology 3. Physics & Materials 4. Control Theory 5. What else that no oneâs thought of yet of yet? © Shutterstock, Inc. Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 13 Neighboring Technologies: Comms ⢠Everything communicates â Radio, networks, fiber â Once, faster processors would win; now designers need to know end applications â Comms theory, signals, modulation techniques, propagation, noise, regulations ⢠All information is probabilistic, with some error rate ⢠Make system accomplish its mission anyway © dalomismo.net Neighboring Technologies: Physics/Bio ⢠Physics & Materials â Photonics, lasers, infrared â Materials (phase - change, flash) â MEMS â Whatever devices follow CMOS ⢠Electronics at the meso and atomic scale ⢠Bio, genetics, human anatomy â Engineered bio â Brain/human interfacing NO Maybe © Royal Society of Chemistry © Reuters/University of Pittsburgh Medical Centre Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 15 In the End: Follow the Money. End of aooreâs Law revives special purpose designs But please! Heed lessons of past. Donât design unprogrammable engines! When Mooreâs Law ends, it will be economics that stops it, not physics. Keep your eye on the money⦠© Shutterstock, Inc. Distribution Statement "A" (Approved for Public Release, Distribution Unlimited) 16 The End © Dill Pixels © XKCD.com Distribution Statement "A" (Approved for Public Release, Distribution Unlimited)