Design and Architectures for Embedded Systems Prof. Dr. J. Henkel - PDF document

Design and Architectures for Embedded Systems Prof. Dr. J. Henkel Henkel Prof. Dr. J. CES CES - - Chair for Embedded Systems Chair for Embedded Systems University of University of Karlsruhe Karlsruhe, Germany , Germany Today: Introduction and Overview Today: Introduction and Overview http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Embedded Systems Examples � � Automotive Automotive ABC, ABS, DSC, … ABC, ABS, DSC, … � � Banking&Money transfer Banking&Money transfer smart cards, … smart cards, … � � Consumer Consumer cell phone, DVD player, MP3 cell phone, DVD player, MP3 player, PDA, … player, PDA, … � Clothing � Clothing electronic textiles electronic textiles � � Environment Environment sensor networks sensor networks � � Healthcare Healthcare hearings aids, pace maker, … hearings aids, pace maker, … � Home Appliances � Home Appliances microwave oven, microwave oven, dishwasher, … dishwasher, … � � Home Automation Home Automation EIB, X10, … EIB, X10, … � � Office Automation Office Automation fax, printer, … fax, printer, … � Security � Security screening, surveillance systems, … screening, surveillance systems, … � � Telecom Systems Telecom Systems satellite, … satellite, … � Test and Measurement Equipment � Test and Measurement Equipment Ampere/Volt- Ampere/Volt -meter, Logic analyzer, … meter, Logic analyzer, … http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004

Example 1 for an Embedded System: Settop Box Chip � � System System- -On On- -a a- -Chip: Chip: uP uP, , caches, ASIC, main caches, ASIC, main memory, I/O memory, I/O � � Today’s Technology: Today’s Technology: � 0.18u … 0.07u 0.18u … 0.07u - -> >> > >> MPEG 100 million 100 million Macro transistors/chip possible transistors/chip possible (core) � Practice: < 100million � Practice: < 100million transistors transistors Cache � why ? � why ? uP -> “productivity gap” > “productivity gap” - (source:NEC Japan) http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Example 2: E-Textiles - Smart Shirt - ] Source: [Marc03] http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004

Example 3: Medical Diagnostics (source: Jan Madsen DTU) http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Example 4: Sensor Networks Disaster Prevention & Energy-efficient Manufacturing plants & Power distribution Emergency buildings • Improve reliability, operating efficiency Response • $55 B / year opportunity in the US Health care • Unwired operating “Smart” environments Traffic control rooms • Homes, Offices, Schools, … • Reduce commute time • Early detection of • Convenience, Productivity, Security by 15 min => $15B/year cardiac attacks in California alone (source: A. Raghunathan, NEC) http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004

Embedded System Definition � There is no concise definition of an embedded system � There is no concise definition of an embedded system � But here are some common characteristics … � But here are some common characteristics … � specialized to an application domain, single application or spec � specialized to an application domain, single application or specific ific task (=> less flexible but probably more efficiently to design) task (=> less flexible but probably more efficiently to design) � Underlies many and tight constraints (later) � Underlies many and tight constraints (later) � Designing an embedded system is typically more challenging than � Designing an embedded system is typically more challenging than designing a general- designing a general -purpose computer purpose computer � � Interacts with the real world => designing embedded software Interacts with the real world => designing embedded software requires the consideration of more constraints than designing requires the consideration of more constraints than designing software for general- -purpose computers purpose computers software for general � By far higher volume than general � By far higher volume than general- -purpose computers. Example: a purpose computers. Example: a single luxury car has already >100 embedded systems single luxury car has already >100 embedded systems � � Market revenue for ES is by far higher than it is for general Market revenue for ES is by far higher than it is for general- - purpose computers (billions instead of millions) purpose computers (billions instead of millions) � New application areas evolve each day … � New application areas evolve each day … http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Moore’s Law � � Gordon E. Moore (co Gordon E. Moore (co- -founded Intel in 1968) founded Intel in 1968) (source: Intel) � Prediction in 1965: � Prediction in 1965: � Exponential growth of number of transistors per � Exponential growth of number of transistors per chip chip Jan. 1971: Intel's � Initial observation: Complexity per die area will 4004 chip; 2,200 � Initial observation: Complexity per die area will transistors; a 4-bit double about every year microprocessor; it double about every year addressed 9.2 K of � Was later relaxed to “doubling every 18 months” � Was later relaxed to “doubling every 18 months” memory � This trend will continue for foreseeable future � This trend will continue for foreseeable future � Original paper: � Original paper: “ “Cramming more components onto integrated circuits”, Electronics, Volume 38, Number 8, April 19, 1965 � 2003: � 2003: “ “ No exponential is forever No exponential is forever … … but we can delay but we can delay ‘ ‘FOREVER FOREVER’ ’ ” ”, talk , talk by G. E. Moore at International Solid State Circuits Conference International Solid State Circuits Conference by G. E. Moore at (ISSCC), Feb. 2003. , Feb. 2003. (ISSCC) � � For more info: International Technology Roadmap for Semiconducto For more info: International Technology Roadmap for Semiconductors: rs: http://public.itrs http://public. itrs.net .net http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004

Visualization of Moore’s Law Factor 2^22 Factor 2^22 � 1970: 1k DRAM (introduced by Intel) � 1970: 1k DRAM (introduced by Intel) i.e. 4 million i.e. 4 million � � 2005: 4G DRAM (150nm) 2005: 4G DRAM (150nm) � 2010: 64G DRAM (80nm) � 2010: 64G DRAM (80nm) 2005 2005 1970 1970 less than credit card size less than credit card size soccer field soccer field size size � Note: exponential growth is larger than many people conceive! � Note: exponential growth is larger than many people conceive! http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Design Productivity Gap Transistors / Chip (w/o memory) Transistors / Staff-month 100000000 10000000 Gap 10000000 Limit to adding more skilled Engineers 1000000 100000 100000 Moore’s Law 10000 1000 1000 Engineering Productivity Trend 100 10 1981 1985 1989 1993 1997 2001 2005 2009 (source: [KeBri98] http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004

Complexity of Microprocessors (source: Intel) http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004 Trends: Crisis of Complexity � Prediction for the case no ESLTools will be used Millions of Gates � However: red curve will 300 apply And lead to SoCs with Available Gates 100s –1000s of PEs per chip 250 Used Gates 200 150 Design Productivity Gap 100 32 50 25 20 10 0.8 1 2 8 43 47 50 55 0.2 0.3 0.4 3 0 2006 1990 1992 1994 1996 1998 2000 2002 2004 [source: Gartner/Dataquest] http://ces.univ-karlsruhe.de J. Henkel, Univ. of Karlsruhe, SS 2004