FAU

Fau-Free PDF

  • Date:23 Nov 2020
  • Views:20
  • Downloads:0
  • Pages:161
  • Size:4.72 MB

Share Pdf : Fau

Download and Preview : Fau


Report CopyRight/DMCA Form For : Fau


Transcription:

Transregional Collaborative Research Centre 89,Invasive Computing. Friedrich Alexander Universit t Erlangen N rnberg,Karlsruher Institut f r Technologie. Technische Universit t M nchen,Annual Report 2016,Coordinator. Prof Dr Ing J rgen Teich,Lehrstuhl f r Informatik 12. Universit t Erlangen N rnberg,Cauerstra e 11 91058 Erlangen.
Managing Director Deputy Managing Director,Dr Ing J rgen Klein der Dr Katja Lohmann. Lehrstuhl f r Informatik 4 Lehrstuhl f r Informatik 12. Universit t Erlangen N rnberg Universit t Erlangen N rnberg. Martensstra e 1 91058 Erlangen Cauerstra e 11 91058 Erlangen. Public Relations,Dr Sandra Mattauch,Lehrstuhl f r Informatik 12. Universit t Erlangen N rnberg,Cauerstra e 11 91058 Erlangen. This report summarises the activities and scientific progress of the. Transregional Collaborative Research Centre 89 Invasive Computing. InvasIC in 2016, The CRC Transregio InvasIC is funded by the Deutsche Forschungs. gemeinschaft in its second funding phase from July 2014 July 2018. The research association aggregates about 60 of the best researchers. from three excellent sites in Germany Friedrich Alexander Universit t. Erlangen N rnberg Karlsruher Institut f r Technologie Technische Uni. versit t M nchen This scientific team includes specialists in algorithm. engineering for parallel algorithm design hardware architects for recon. figurable MPSoC development as well as language tool and application. and operating system designers, Exciting events in 2016 certainly were two Dagstuhl Seminars co.
organised by members of our CRC Transregio In January Prof Dr. Michael Gerndt TUM and Prof Dr Ing Michael Gla FAU organised. and coordinated the Seminar 16052 on Dark Silicon From Embedded. to HPC Systems Prof Dr Ing J rgen Teich FAU and Prof Dr Ir Ingrid. Verbauwhede KU Leuven BE were the organisers of the Seminar 16441. Adaptive Isolation for Predictability and Security which took place in. 2016 also was an outstanding year concerning the professional success. of three members of the CRC Transregio Prof Dr Ing Michael Gla. FAU Dr Ing Muhammad Shafique KIT and PD Dr Ing habil Daniel. Lohmann FAU were appointed as professors at the highly reputable. universities of Ulm Vienna and Hanover respectively. We would like to thank all members of the CRC Transregio InvasIC. and all our partners from industry and academia for the fruitful collabo. rations and inspiring discussions in the last year We do hope that you. will enjoy reading about the progress achieved in 2016 as well as about. our research planned for the following years,J rgen Teich. Coordinator,Contents 4,I Invasive Computing 7,1 About InvasIC 8. 2 Participating University Groups 11,II Research Program 13. 3 Overview of Research Program 14,4 Research Projects 17. A1 Basics of Invasive Computing 17,A4 Design Time Characterisation and Analysis.
of Invasive Algorithmic Patterns 27,B1 Adaptive Application Specific. Invasive Microarchitecture 35,B2 Invasive Tightly coupled Processor Arrays 41. B3 Power Efficient Invasive Loosely Coupled MPSoCs 49. B4 Hardware Monitoring System and Design,Optimisation for Invasive Architectures 55. B5 Invasive NoCs Autonomous,Self Optimising Communication. Infrastructures for MPSoCs 58,C1 Invasive Run Time Support System iRTSS 66.
C2 Simulative Design Space Exploration 77,C3 Compilation and Code Generation. for Invasive Programs 83,C5 Security in Invasive Computing Systems 92. D1 Invasive Software Hardware Architectures for Robotics 98. D3 Invasion for High Performance Computing 104,Z Central Services 110. Z2 Validation and Demonstrator 111,5 Working Groups 117. WG1 Predictability 117,WG2 Memory Hierarchy 121,WG3 Benchmarking and Evaluation 124.
WG4 Power Efficiency and Dark Silicon 126,III Events and Activities 131. 6 Internal Meetings 133,7 Training Courses and Tutorials 134. 8 InvasIC Activities 137,9 Industrial and Scientific Board 149. 10 Publications 151,Invasive Computing,1 About InvasIC. The Idea of Invasive Computing, The main idea and novelty of invasive computing is to introduce resource.
aware programming support in the sense that a given program gets the. ability to explore and dynamically spread its computations to processors. similar to a phase of invasion then to execute portions of code of high. parallelism degree in parallel based on the available invasible region. on a given multi processor architecture Afterwards once the program. terminates or if the degree of parallelism should be lower again the. program may enter a retreat phase deallocate resources and resume. execution again for example sequentially on a single processor To. support this idea of self adaptive and resource aware programming. new programming concepts languages compilers and operating sys. tems are necessary as well as architectural changes in the design of. MPSoCs Multi Processor Systems on a Chip to efficiently support inva. sion infection and retreat operations by involving concepts for dynamic. processor interconnect and memory reconfiguration Decreasing fea. ture sizes have also led to a rethinking in the design of multi million. transistor system on chip SoC architectures envisioning dramatically. increasing rates of temporary and permanent faults and feature varia. As we can foresee SoCs with 1000 or more processors on a single chip. in the year 2020 static and central management concepts to control the. execution of all resources might have met their limits long before and. are therefore not appropriate Invasion might provide the required self. organising behaviour to conventional programs for being able to provide. scalability higher resource utilisation required fault tolerance and of. course also performance gains by adjusting the amount of allocated. resources to the temporal needs of a running application This thought. opens a new way of thinking about parallel algorithm design Based. on algorithms utilising invasion and negotiating resources with others. we can imagine that corresponding programs become personalised ob. jects competing with other applications running simultaneously on an. First Achievements, A Transregional Collaborative Research Centre aggregating the best. researchers from three excellent sites in Germany provides an ideal base. to investigate the above revolutionary ideas Starting off at basically. zero in terms of invasive processor hardware language compiler and. operating system availability we have truly fostered the fundamentals of. invasive computing in our first funding phase These include the defini. tion of required programming language elements for the specification of. invasion operations as well as a set of constraints to argue about number. types and state of resources that may be invaded defining the invasive. command space Project Area A A first invasive language based on the. language X10 by IBM as well as a compiler for translation of invasive. X10 programs Project Area C onto a heterogeneous invasive multi tile. architecture that has also been successfully jointly architected Project. Area B is meanwhile ready for experimentation on an FPGA based. prototype Project Z2 The compiler interfaces to the invasive run. time support system iRTSS that provides for dedicated operating system. support for invasive computing First invasive applications exploiting. different types of processor and communication resources of an inva. sive network on chip iNoC are running successfully and have shown. considerable gains in resource utilisation and computational efficiencies. in comparison with their non invasive counterparts. Current Scientific Goals, A unique jewel of invasive computing however has not been exploited. at all so far By the fact that resources are temporally claimed by. default in an exclusive manner interferences due to multiple applica. tions sharing the same resources being the reality on today s multicore. systems may be reduced if not avoided completely Moreover run to. completion is the default mode of thread execution Finally memory. reconfiguration and isolation as well as bandwidth guarantees on the. designed network on chip allow us also to provide predictable QoS also. for communication, In the current funding phase we play out this ace systematically. by tackling a predictability of b multi objective execution qualities. of parallel invasive programs and including their c optimisation and. exploration of design space Our joint investigations include new lan. guage constructs to define so called requirements on desired respectively. amended qualities of execution Application specified qualities may not. only be of type performance e g execution time throughput etc. but do also include aspects of security and fault tolerance Through. analysis of application requirements from different domains including. stream processing and malleable task applications not only efficiency. but also predictable execution qualities shall be demonstrated for appli. cations stemming from robotics imaging as well as HPC As another. new yet very important facet of invasive computing a special focus in. the current funding phase is devoted to the problem of dark silicon and. energy efficient computing,Long Term Vision, With the aforementioned fundamental investigations in mind we intend. to demonstrate that invasive computing will be a if not the vehicle. for solving many current problems of multicore computing today by. providing resource awareness for a mixture of best effort applications. and applications with predictable quality We do expect that a huge. application and business field in embedded system applications might. be accessed through the foundations of invasive computing. 2 Participating University Groups,Friedrich Alexander Universit t Erlangen N rnberg.
Lehrstuhl f r Hardware Software Co Design,Prof Dr Ing J rgen Teich. Prof Dr Ing Michael Gla,Dr Ing Frank Hannig,Dr Ing Stefan Wildermann. Lehrstuhl f r IT Sicherheitsinfrastrukturen,Prof Dr Ing Felix Freiling. Lehrstuhl f r Verteilte Systeme und Betriebssysteme. Prof Dr Ing Wolfgang Schr der Preikschat,PD Dr Ing Daniel Lohmann. Karlsruher Institut f r Technologie,Institut f r Anthropomatik und Robotik.
Prof Dr Ing Tamim Asfour, Institut f r Programmstrukturen und Datenorganisation. Prof Dr Ing Gregor Snelting,Institut f r Technik der Informationsverarbeitung. Prof Dr Ing J rgen Becker,Institut f r Technische Informatik. Prof Dr Ing J rg Henkel,Dr Ing Lars Bauer,Technische Universit t M nchen. Lehrstuhl f r Entwurfsautomatisierung,Prof Dr Ing Ulf Schlichtmann.
Lehrstuhl f r integrierte Systeme,Prof Dr sc techn Andreas Herkersdorf. Prof Dr Ing Walter Stechele,Dr Ing Thomas Wild, Lehrstuhl f r Rechnertechnik und Rechnerorganisation. Prof Dr Michael Gerndt,Lehrstuhl f r Technische Elektronik. Prof Dr rer nat Doris Schmitt Landsiedel,Lehrstuhl f r Wissenschaftliches Rechnen. Prof Dr Hans Joachim Bungartz,Prof Dr Michael Bader.
Research Program,3 Overview of Research Program, To investigate the main aspects of invasive computing the CRC Tran. sregio is organised in five project areas, Area A Fundamentals Language and Algorithm Research. Research in project area A focuses on the basic concepts of invasion and. resource aware programming as well as on language issues algorithmic. theory of invasion and on analysis and optimisation techniques for. application characterisation and hybrid mixed static dynamic core. allocation,Area B Architectural Research, Project area B investigates micro and macroarchitectural requirements. techniques and hardware concepts to enable invasive computing in. future MPSoCs,Area C Compiler Simulation and Run Time Support. The focus of project area C is on software support for invasive computing. including compiler simulation and operating system functionality as. well as on design space exploration with a special focus on run time. management,Area D Applications, Applications serve as demonstrators for the diverse and efficient de.
ployment of invasive computing The applications have been chosen. carefully from the domains of robotics and scientific computing in order. to demonstrate distinct and complementary features of invasive comput. ing for example its capability to provide quality predictable execution. of parallel programs,Z2 Validation and Demonstrator. A hardware demonstrator will serve again as the key concept for vali. dation of invasive computing principles It will allow for co validation. and demonstration of invasive computing through tight integration of. hardware and software research results at the end of the second project. phase and to decide on the further roadmap of specific hardware for. invasive computing, The four working groups Predictability Memory Hierarchy Bench. marking and Evaluation and Power Efficiency and Dark Silicon de. fined on top of these project areas support the interdisciplinary research. Coordinator Prof Dr Ing J rgen Teich Lehrstuhl f r Informatik 12 Universit t Erlangen N rnberg Cauerstra e 11 91058 Erlangen Managing Director Dr Ing J rgen Klein der

Related Books