Design Space Exploration of Embedded Systems © Lothar Thiele ETH Zurich Overview

Design Space Exploration of Embedded Systems © Lothar Thiele ETH Zurich Overview www.phwiki.com

Design Space Exploration of Embedded Systems © Lothar Thiele ETH Zurich Overview

Fonseca, Felicia, Northern Arizona Correspondent has reference to this Academic Journal, PHwiki organized this Journal Design Space Exploration of Embedded Systems © Lothar Thiele ETH Zurich Overview Review of General Aspects Basic Models in addition to Methods Modular Per as long as mance Analysis Multi-Criteria Optimization Applications Concluding Remarks Target Plat as long as m Heterogeneous computing in addition to memory resources some resource types: GP processors, ASIPs (DSP, micro-controller), weakly programmable co-processors, re-configurable components, hard coded IP components heterogeneous plat as long as m software: RTOS, scheduling (pre-emptive, static, dynamic), synchronization DSP mC image coprocessor CAN interface SDRAM RISC FPGA

ITT Technical Institute Indianapolis US www.phwiki.com

This Particular University is Related to this Particular Journal

Target Plat as long as m Communication micro-network on chip as long as synchronization in addition to data exchange consisting of busses, routers, drivers some critical issues: topology, switching strategies (packet, circuit), routing strategies (static – reconfigurable – dynamic), arbitration policies (dynamic, TDM, CDMA, fixed priority) challenges: heterogeneous components in addition to requirements, compose network that matches the traffic characteristics of a given application (domain) Design Space TDMA Priority EDF WFQ RISC DSP LookUp Cipher mE mE mE mE mE mE static Design Space Exploration

Design Space Exploration Design Space Exploration (Semi-) Automated Design Space Exploration Why is Per as long as mance Analysis Difficult complex behavior – input stream – data dependent behavior I/O DSP CPU1 CPU2 task1 task2 task3 task4 interference – limited resources – scheduling/arbitration interference of multiple applications – limited resources – scheduling/arbitration – anomalies

Simulation Target architecture co-simulation combines functional in addition to per as long as mance validation extensive runtimes worst case inputs test case definition re-targeting expensive input trace mixed model function: application structure: hardware-software architecture output trace Trace-Based Simulation Steps: execution trace determined by co-simulation abstract representation using communication graph extension of graph by actual architecture Faster than simulation, but still based on single trace input trace functional model complete trace communication architecture abstract graph trace simulation estimation results [Lahiri et. al, 2001] Static Analytic Models Steps: describe computing, communication in addition to memory resources by algebraic equations, e.g. describe properties of input using parameters, e.g. input data rate combine relations Fast in addition to simple estimation Generally inaccurate modeling of shared resources

Dynamic Analytic Models Combination between static models, possibly extended by their dynamic behavior, e.g. non-determinism in run-time in addition to event processing dynamic models as long as describing shared resources (scheduling in addition to arbitration) Variants queuing theory (statistical models, average case) real-time calculus (interval methods, worst case) More accurate than static models Dynamic Analytic Models input traces model of environment spec. of inputs component simulation system model estimation results data sheets model of components model of architecture Summary Simulation Trace-based simulation Dynamic analytic methods Static analytic methods Timing Accuracy Run-time

Bounds, Guarantees in addition to Predictability Example: end-to-end delay design Overview Review of General Aspects Basic Models in addition to Methods Modular Per as long as mance Analysis Multi-Criteria Optimization Applications Concluding Remarks Examples Event Stream Processing

Application Model Example of a simple stream processing task structure: Architecture Templates In general, we assume an arbitrary heterogeneous architecture consisting of computing resources, memory in addition to communication resources. event event events Mapping Model

Abstraction Idea: unified view of task scheduling, arbitration in addition to event scheduling in networks: methods: queueing theory (statistical bounds, markov chains) real-time calculus (worst case bounds, min-max algebra) Real-Time Calculus Example of a dynamic analytic model Characteristics yields worst case estimation results as long as memory, delay, throughput takes into account application structure (task graph representation, SPI) architecture in addition to mapping (computation, communication, scheduling) environment (characterization of input traces) Foundations of Real-Time-Calculus Linear System Theory [Baccelli, Cohen, Olsder, Quadrat 1992] Calculus as long as Networks [Le Boudec 1998, 2001], [Cruz 1991] Adversarial Queuing Theory [Andrews, Borodin, Kleinberg, Leighton, 1996] Competing Approaches: SymTA/S: [R. Ernst et. al. 2002] Holistic Scheduling: [P. Eles 1999] Distributed Scheduling: [Tindell, Burns 1996] Recurring Task Models: [S. Baruah, 1998, 2002]

Fonseca, Felicia Associated Press (AP) - Phoenix Bureau Northern Arizona Correspondent www.phwiki.com

Overview Review of General Aspects Basic Models in addition to Methods Modular Per as long as mance Analysis Multi-Criteria Optimization Applications Concluding Remarks Elements of Modular Per as long as mance Analysis system architecture model architectural element model environment model per as long as mance model mapping, scheduling application hardware architecture analysis Event Streams (Environment Model) How do we model uncertain event streams time t [ms] Event Stream

Event Streams (Environment Model) Use interval bound functions: Arrival Curves time t [ms] Event Stream Event Streams (Environment Model) Use interval bound functions: Arrival Curves max: 1 event min: 0 events max: 2 events min: 0 events max: 3 events min: 1 events time t [ms] D [ms] 0 1 2 of events 1 2 3 au al Event Stream Arrival Curves [al, au] Elements of Modular Per as long as mance Analysis system architecture model architectural element model environment model per as long as mance model mapping, scheduling application hardware architecture analysis

Concluding Remarks SystemC SystemC Acknowledgement in addition to References The presentation contains contributions by Samarjit Chakraborty (NUS) Simon Künzli, Ernesto W in addition to eler, Alex in addition to er Maxiaguine (ETHZ) Andreas Herkersdorf, Patricia Sagmeister (IBM) Jonas Greutert (Netmodule) Many publications are available from http://www.tik.ee.ethz.ch/~thiele

Fonseca, Felicia Northern Arizona Correspondent

Fonseca, Felicia is from United States and they belong to Associated Press (AP) – Phoenix Bureau and they are from  Flagstaff, United States got related to this Particular Journal. and Fonseca, Felicia deal with the subjects like Regional News

Journal Ratings by ITT Technical Institute Indianapolis

This Particular Journal got reviewed and rated by ITT Technical Institute Indianapolis and short form of this particular Institution is US and gave this Journal an Excellent Rating.