Es wird ein Ansatz zur zeitbeschränkten Ablaufplanung für VLIW-Prozessoren mit neuronalen Netzen vorgestellt. Bestehende Arbeiten werden dahingehend erweitert, dass der Datenpfad des Prozessors über heterogene funktionale Einheiten verfügen und geclustert sein darf. Es werden zwei Varianten zur Lösung des Problems angegeben, deren Qualität mit einem heuristischen Ansatz verglichen wird und Schlussfolgerungen bezüglich der Nutzbarkeit neuronaler Netze gezogen.

Der aktuelle Stand der Technologie bringt schnellere und kleinere Bausteine für die Herstellung von integrierten Schaltungen mit sich, die während sie effizienter sind auch anfälliger für Strahlung werden. Kleinere Abmessungen der Transistoren, höhere Integrationsdichte, geringere Versorgungsspannungen und höhere Betriebsfrequenzen sind einige der Charakteristika, die energiegeladene Partikel zu einer Herausforderung machen, wenn man integrierte Schaltungen in rauen Umgebungen einsetzt. Diese Art der Partikel hat einen sehr großen Einfluss auf Prozessoren, die in einer solchen Umgebung eingesetzt werden. Sowohl die Ausführung des Programms, welche durch fehlerhafte Sprünge in der Programmsequenz beeinflusst wird, als auch Daten, die in speichernden Elementen wie Programmspeicher, Datenspeicher oder in Registern abgelegt sind, werden verfälscht. Um solche Prozessorsysteme abzusichern, wird in der Literatur Fehlertoleranz empfohlen, welche die Systemperformanz verringert, einen größeren Flächenverbrauch mit sich bringt und das System dennoch nicht komplett schützen kann. Diese Fehlertoleranz kann sowohl durch software- als auch durch hardwarebasierte Ansätze umgesetzt werden. In diesem Zusammenhang schlagen wir eine Kombination aus Hardware- und Software- Lösung vor, welche die Systemperformanz nur sehr wenig beeinflusst und den zusätzlichen Speicheraufwand minimiert. Diese Hybrid-Technologie zielt darauf ab, alle Fehler in einem System zu finden. Fünf solcher Techniken werden beschrieben und erklärt, zwei der vorgestellten Techniken sind bekannte Software-Lösungen, die anderen drei sind neue Hybrid-Lösungen, um alle transienten Effekte von Strahlung in Prozessoren erkennen zu können. Diese unterschiedlichen Ansätze werden anhand ihrer Ausführungszeit, Programm-, Datenspeicher, Flächenvergrößerung und Taktfrequenz analysiert und ausgewertet. Um die Effizienz und die Machbarkeit des vorgeschlagenen Ansatzes verifizieren zu können, werden Fehlerinjektionstests sowohl durch Simulation als auch durch Bestrahlungsexperimente in unterschiedlichen Positionen mit einer Cobalt-60 Quelle durchgeführt. Die Ergebnisse des vorgeschlagenen Ansatzes verbessern den Stand der Technik durch die Bereitstellung einer höheren Fehlererkennungsrate bei sehr geringer negativer Beeinflussung der Performanz und des Speicherverbrauchs. / Os recentes avanços tecnológicos proporcionaram dispositivos menores e mais rápidos para a fabricação de circuitos que, apesar de mais eficientes, se tornaram mais sensíveis aos efeitos de radiação. Menores dimensões de transistores, mais densidade de integração, tensões de alimentação mais baixas e frequências de operação mais altas são algumas das características que tornaram partículas energizadas um problema, quando lidando com sistemas integrados em ambientes severos. Estes tipos de partículas tem uma grande influencia em processadores funcionando em tais ambientes, afetando tanto o fluxo de execução do programa ao causar desvios incorretos, bem como os dados armazenados em elementos de memória, como memórias de dados e programas e registradores. A fim de proteger sistemas processados, técnicas de tolerância a falhas foram propostas na literatura usando propostas baseadas em hardware, software, que diminuem o desempenho do sistema, aumentam a sua área e não são capazes de proteger totalmente o sistema destes efeitos. Neste contexto, propomos a combinação de técnicas baseadas em hardware e software para criar técnicas híbridas orientadas a detectar todas as falhas que afetam o sistema, com baixa degradação de desempenho e aumento de memória. Cinco técnicas são apresentadas e descritas em detalhes, das quais duas são conhecidas técnicas baseadas puramente em software e três são técnicas híbridas novas, para detectar todos os tipos de efeitos transientes causados pela radiação em processadores. As técnicas são avaliadas de acordo com o aumento no tempo de execução, no uso das memórias de dados e programa e de área, e degradação da frequência de operação. Para verificar a eficiência e aplicabilidade das técnicas propostas, campanhas de injeção de falhas são realizadas ao se simular a injeção de falhas e realizar experimentos de irradiação em diferentes localidades com nêutron e fontes de Cobalto-60. Os resultados mostraram que as técnicas propostas aprimoraram o estado da arte ao fornecer altas taxas de detecção de falhas com baixas penalidades em degradação de desempenho e aumento de memória. / Recent technology advances have provided faster and smaller devices for manufacturing circuits that while more efficient have become more sensitive to the effects of radiation. Smaller transistor dimensions, higher density integration, lower voltage supplies and higher operating frequencies are some of the characteristics that make energized particles an issue when dealing with integrated circuits in harsh environments. These types of particles have a major influence in processors working in such environments, affecting both the program’s execution flow by causing incorrect jumps in the program, and the data stored in memory elements, such as data and program memories, and registers. In order to protect processor systems, fault tolerance techniques have been proposed in literature using hardware-based and software-based approaches, which decrease the system’s performance, increase its area, and are not able to fully protect the system against such effects. In this context, we proposed a combination of hardware- and software-based techniques to create hybrid techniques aimed at detecting all the faults affecting the system, at low performance degradation and memory overhead. Five techniques are presented and described in detail, from which two are known software-based only techniques and three are new hybrid techniques, to detect all kinds of transient effects caused by radiation in processors. The techniques are evaluated according to execution time, program and data memories, and area overhead and operating frequency degradation. To verify the effectiveness and the feasibility of the proposed techniques, fault injection campaigns are performed by injecting faults by simulation and performing irradiation experiments in different locations with neutrons and a Cobalt-60 sources. Results have shown that the proposed techniques improve the state-of-the-art by providing high fault detection rates at low penalties on performance degradation and memory overhead.

Der aktuelle Stand der Technologie bringt schnellere und kleinere Bausteine für die Herstellung von integrierten Schaltungen mit sich, die während sie effizienter sind auch anfälliger für Strahlung werden. Kleinere Abmessungen der Transistoren, höhere Integrationsdichte, geringere Versorgungsspannungen und höhere Betriebsfrequenzen sind einige der Charakteristika, die energiegeladene Partikel zu einer Herausforderung machen, wenn man integrierte Schaltungen in rauen Umgebungen einsetzt. Diese Art der Partikel hat einen sehr großen Einfluss auf Prozessoren, die in einer solchen Umgebung eingesetzt werden. Sowohl die Ausführung des Programms, welche durch fehlerhafte Sprünge in der Programmsequenz beeinflusst wird, als auch Daten, die in speichernden Elementen wie Programmspeicher, Datenspeicher oder in Registern abgelegt sind, werden verfälscht. Um solche Prozessorsysteme abzusichern, wird in der Literatur Fehlertoleranz empfohlen, welche die Systemperformanz verringert, einen größeren Flächenverbrauch mit sich bringt und das System dennoch nicht komplett schützen kann. Diese Fehlertoleranz kann sowohl durch software- als auch durch hardwarebasierte Ansätze umgesetzt werden. In diesem Zusammenhang schlagen wir eine Kombination aus Hardware- und Software- Lösung vor, welche die Systemperformanz nur sehr wenig beeinflusst und den zusätzlichen Speicheraufwand minimiert. Diese Hybrid-Technologie zielt darauf ab, alle Fehler in einem System zu finden. Fünf solcher Techniken werden beschrieben und erklärt, zwei der vorgestellten Techniken sind bekannte Software-Lösungen, die anderen drei sind neue Hybrid-Lösungen, um alle transienten Effekte von Strahlung in Prozessoren erkennen zu können. Diese unterschiedlichen Ansätze werden anhand ihrer Ausführungszeit, Programm-, Datenspeicher, Flächenvergrößerung und Taktfrequenz analysiert und ausgewertet. Um die Effizienz und die Machbarkeit des vorgeschlagenen Ansatzes verifizieren zu können, werden Fehlerinjektionstests sowohl durch Simulation als auch durch Bestrahlungsexperimente in unterschiedlichen Positionen mit einer Cobalt-60 Quelle durchgeführt. Die Ergebnisse des vorgeschlagenen Ansatzes verbessern den Stand der Technik durch die Bereitstellung einer höheren Fehlererkennungsrate bei sehr geringer negativer Beeinflussung der Performanz und des Speicherverbrauchs. / Os recentes avanços tecnológicos proporcionaram dispositivos menores e mais rápidos para a fabricação de circuitos que, apesar de mais eficientes, se tornaram mais sensíveis aos efeitos de radiação. Menores dimensões de transistores, mais densidade de integração, tensões de alimentação mais baixas e frequências de operação mais altas são algumas das características que tornaram partículas energizadas um problema, quando lidando com sistemas integrados em ambientes severos. Estes tipos de partículas tem uma grande influencia em processadores funcionando em tais ambientes, afetando tanto o fluxo de execução do programa ao causar desvios incorretos, bem como os dados armazenados em elementos de memória, como memórias de dados e programas e registradores. A fim de proteger sistemas processados, técnicas de tolerância a falhas foram propostas na literatura usando propostas baseadas em hardware, software, que diminuem o desempenho do sistema, aumentam a sua área e não são capazes de proteger totalmente o sistema destes efeitos. Neste contexto, propomos a combinação de técnicas baseadas em hardware e software para criar técnicas híbridas orientadas a detectar todas as falhas que afetam o sistema, com baixa degradação de desempenho e aumento de memória. Cinco técnicas são apresentadas e descritas em detalhes, das quais duas são conhecidas técnicas baseadas puramente em software e três são técnicas híbridas novas, para detectar todos os tipos de efeitos transientes causados pela radiação em processadores. As técnicas são avaliadas de acordo com o aumento no tempo de execução, no uso das memórias de dados e programa e de área, e degradação da frequência de operação. Para verificar a eficiência e aplicabilidade das técnicas propostas, campanhas de injeção de falhas são realizadas ao se simular a injeção de falhas e realizar experimentos de irradiação em diferentes localidades com nêutron e fontes de Cobalto-60. Os resultados mostraram que as técnicas propostas aprimoraram o estado da arte ao fornecer altas taxas de detecção de falhas com baixas penalidades em degradação de desempenho e aumento de memória. / Recent technology advances have provided faster and smaller devices for manufacturing circuits that while more efficient have become more sensitive to the effects of radiation. Smaller transistor dimensions, higher density integration, lower voltage supplies and higher operating frequencies are some of the characteristics that make energized particles an issue when dealing with integrated circuits in harsh environments. These types of particles have a major influence in processors working in such environments, affecting both the program’s execution flow by causing incorrect jumps in the program, and the data stored in memory elements, such as data and program memories, and registers. In order to protect processor systems, fault tolerance techniques have been proposed in literature using hardware-based and software-based approaches, which decrease the system’s performance, increase its area, and are not able to fully protect the system against such effects. In this context, we proposed a combination of hardware- and software-based techniques to create hybrid techniques aimed at detecting all the faults affecting the system, at low performance degradation and memory overhead. Five techniques are presented and described in detail, from which two are known software-based only techniques and three are new hybrid techniques, to detect all kinds of transient effects caused by radiation in processors. The techniques are evaluated according to execution time, program and data memories, and area overhead and operating frequency degradation. To verify the effectiveness and the feasibility of the proposed techniques, fault injection campaigns are performed by injecting faults by simulation and performing irradiation experiments in different locations with neutrons and a Cobalt-60 sources. Results have shown that the proposed techniques improve the state-of-the-art by providing high fault detection rates at low penalties on performance degradation and memory overhead.

Der aktuelle Stand der Technologie bringt schnellere und kleinere Bausteine für die Herstellung von integrierten Schaltungen mit sich, die während sie effizienter sind auch anfälliger für Strahlung werden. Kleinere Abmessungen der Transistoren, höhere Integrationsdichte, geringere Versorgungsspannungen und höhere Betriebsfrequenzen sind einige der Charakteristika, die energiegeladene Partikel zu einer Herausforderung machen, wenn man integrierte Schaltungen in rauen Umgebungen einsetzt. Diese Art der Partikel hat einen sehr großen Einfluss auf Prozessoren, die in einer solchen Umgebung eingesetzt werden. Sowohl die Ausführung des Programms, welche durch fehlerhafte Sprünge in der Programmsequenz beeinflusst wird, als auch Daten, die in speichernden Elementen wie Programmspeicher, Datenspeicher oder in Registern abgelegt sind, werden verfälscht. Um solche Prozessorsysteme abzusichern, wird in der Literatur Fehlertoleranz empfohlen, welche die Systemperformanz verringert, einen größeren Flächenverbrauch mit sich bringt und das System dennoch nicht komplett schützen kann. Diese Fehlertoleranz kann sowohl durch software- als auch durch hardwarebasierte Ansätze umgesetzt werden. In diesem Zusammenhang schlagen wir eine Kombination aus Hardware- und Software- Lösung vor, welche die Systemperformanz nur sehr wenig beeinflusst und den zusätzlichen Speicheraufwand minimiert. Diese Hybrid-Technologie zielt darauf ab, alle Fehler in einem System zu finden. Fünf solcher Techniken werden beschrieben und erklärt, zwei der vorgestellten Techniken sind bekannte Software-Lösungen, die anderen drei sind neue Hybrid-Lösungen, um alle transienten Effekte von Strahlung in Prozessoren erkennen zu können. Diese unterschiedlichen Ansätze werden anhand ihrer Ausführungszeit, Programm-, Datenspeicher, Flächenvergrößerung und Taktfrequenz analysiert und ausgewertet. Um die Effizienz und die Machbarkeit des vorgeschlagenen Ansatzes verifizieren zu können, werden Fehlerinjektionstests sowohl durch Simulation als auch durch Bestrahlungsexperimente in unterschiedlichen Positionen mit einer Cobalt-60 Quelle durchgeführt. Die Ergebnisse des vorgeschlagenen Ansatzes verbessern den Stand der Technik durch die Bereitstellung einer höheren Fehlererkennungsrate bei sehr geringer negativer Beeinflussung der Performanz und des Speicherverbrauchs. / Os recentes avanços tecnológicos proporcionaram dispositivos menores e mais rápidos para a fabricação de circuitos que, apesar de mais eficientes, se tornaram mais sensíveis aos efeitos de radiação. Menores dimensões de transistores, mais densidade de integração, tensões de alimentação mais baixas e frequências de operação mais altas são algumas das características que tornaram partículas energizadas um problema, quando lidando com sistemas integrados em ambientes severos. Estes tipos de partículas tem uma grande influencia em processadores funcionando em tais ambientes, afetando tanto o fluxo de execução do programa ao causar desvios incorretos, bem como os dados armazenados em elementos de memória, como memórias de dados e programas e registradores. A fim de proteger sistemas processados, técnicas de tolerância a falhas foram propostas na literatura usando propostas baseadas em hardware, software, que diminuem o desempenho do sistema, aumentam a sua área e não são capazes de proteger totalmente o sistema destes efeitos. Neste contexto, propomos a combinação de técnicas baseadas em hardware e software para criar técnicas híbridas orientadas a detectar todas as falhas que afetam o sistema, com baixa degradação de desempenho e aumento de memória. Cinco técnicas são apresentadas e descritas em detalhes, das quais duas são conhecidas técnicas baseadas puramente em software e três são técnicas híbridas novas, para detectar todos os tipos de efeitos transientes causados pela radiação em processadores. As técnicas são avaliadas de acordo com o aumento no tempo de execução, no uso das memórias de dados e programa e de área, e degradação da frequência de operação. Para verificar a eficiência e aplicabilidade das técnicas propostas, campanhas de injeção de falhas são realizadas ao se simular a injeção de falhas e realizar experimentos de irradiação em diferentes localidades com nêutron e fontes de Cobalto-60. Os resultados mostraram que as técnicas propostas aprimoraram o estado da arte ao fornecer altas taxas de detecção de falhas com baixas penalidades em degradação de desempenho e aumento de memória. / Recent technology advances have provided faster and smaller devices for manufacturing circuits that while more efficient have become more sensitive to the effects of radiation. Smaller transistor dimensions, higher density integration, lower voltage supplies and higher operating frequencies are some of the characteristics that make energized particles an issue when dealing with integrated circuits in harsh environments. These types of particles have a major influence in processors working in such environments, affecting both the program’s execution flow by causing incorrect jumps in the program, and the data stored in memory elements, such as data and program memories, and registers. In order to protect processor systems, fault tolerance techniques have been proposed in literature using hardware-based and software-based approaches, which decrease the system’s performance, increase its area, and are not able to fully protect the system against such effects. In this context, we proposed a combination of hardware- and software-based techniques to create hybrid techniques aimed at detecting all the faults affecting the system, at low performance degradation and memory overhead. Five techniques are presented and described in detail, from which two are known software-based only techniques and three are new hybrid techniques, to detect all kinds of transient effects caused by radiation in processors. The techniques are evaluated according to execution time, program and data memories, and area overhead and operating frequency degradation. To verify the effectiveness and the feasibility of the proposed techniques, fault injection campaigns are performed by injecting faults by simulation and performing irradiation experiments in different locations with neutrons and a Cobalt-60 sources. Results have shown that the proposed techniques improve the state-of-the-art by providing high fault detection rates at low penalties on performance degradation and memory overhead.

Improving a computers performance has been of major interest to all users around the world, from computing centers to private persons, ever since computer science has entered the stage and then the spotlight in the 1940’s. Most often times, this is either achieved by exchanging parts of the computer with better performing parts, called an upgrade, or by simply buying a newer and better computer. Another approach, which originates from the scientific community, is the optimization of the source code of an application. Thereby, the application programmer capitalizes his knowledge about the underlying platform and its tool-chain in order to gain tweaked binary code, which results in a better performance. It is clear, that this technique will never be an option for consumer electronics or people outside the area of programming and software development. Traditionally, these users stick with the upgrade/buy new method. During the last years, consumer electronics improved into multi-tool devices, which are capable of almost any functionality, originating from their internet connection and their ability to dynamically download and install new software. Certainly, it may happen that an application is too demanding for a given underlying hardware revision. As these new devices are built in a monolithic way, a hardware upgrade is not an option. Nonetheless, most users do not want to buy a new device every time this happens. Thus, it is necessary to provide a possibility, which allows the processor to adapt to a given application at runtime, and thereby improving its own performance. This thesis presents three major approaches to such a runtime dynamic application acceleration.

Im Laufe der letzten Jahre haben sich die Anwendungsbereiche von Computern immer stärker erweitert. Am Anfang der Entwicklung stand die Verwendung als reine Rechenmaschine im Vordergrund. In der weiteren Entwicklung erfolgte die Verwendung vornehmlich als Eingabe- und Verwaltungsgerät für umfangreiche Datenmengen bzw. zur Prozeßsteuerung. Seit Anfang der 90er Jahre nimmt jedoch die Nutzung als Kommunikations- und Informationsgerät stark zu. In diesem Zusammenhang hat sich auch die Präsentation der vom Computer verarbeiteten Daten stark gewandelt. Statt alphanumerischen Textausgaben werden heute die Daten häufig aufwendig grafisch aufbereitet und dargestellt. Diese Entwicklung ist in erster Linie durch die gesteigerte Leistungsfähigkeit der Hard- und Software (Prozessoren, Speicher, Netze, Betriebssysteme usw.) möglich geworden. Aufgrund dieser Steigerung sind heute handelsübliche Computer in der Lage, Datenmengen zu verarbeiten, die bisher nur Spezialmaschinen vorbehalten waren. Dazu zählen unter anderem Videodaten in einer Qualität, wie sie von Videorecordern aufgezeichnet werden.

Data driven organizations gather information on various aspects of their endeavours and analyze that information to gain valuable insights or to increase automatization. Today, these organizations can choose from a wealth of specialized analytical libraries and platforms to meet their functional and non-functional requirements. Indeed, many common application scenarios involve the combination of multiple such libraries and platforms in order to provide a holistic perspective. Due to the scattered landscape of specialized analytical tools, this integration can result in complex and hard to evolve applications. In addition, the necessary movement of data between tools and formats can introduce a serious performance penalty. In this article we present a unified programming environment for analytical applications. The environment includes AL, a programming language that combines concepts of various common analytical domains. Further, the environment also includes a flexible compilation system that uses a language-, domain-, and platform independent program intermediate representation to separate high level application logic and physical organisation. We provide a detailed introduction of AL, establish our program intermediate representation as a generally useful abstraction, and give a detailed explanation of the translation of AL programs into workloads for our experimental shared-memory processing engine.

In continuation of a successful series of events, the 4th Many-core Applications Research Community (MARC) symposium took place at the HPI in Potsdam on December 8th and 9th 2011. Over 60 researchers from different fields presented their work on many-core hardware architectures, their programming models, and the resulting research questions for the upcoming generation of heterogeneous parallel systems.

Improving a computers performance has been of major interest to all users around the world, from computing centers to private persons, ever since computer science has entered the stage and then the spotlight in the 1940’s. Most often times, this is either achieved by exchanging parts of the computer with better performing parts, called an upgrade, or by simply buying a newer and better computer. Another approach, which originates from the scientific community, is the optimization of the source code of an application. Thereby, the application programmer capitalizes his knowledge about the underlying platform and its tool-chain in order to gain tweaked binary code, which results in a better performance. It is clear, that this technique will never be an option for consumer electronics or people outside the area of programming and software development. Traditionally, these users stick with the upgrade/buy new method. During the last years, consumer electronics improved into multi-tool devices, which are capable of almost any functionality, originating from their internet connection and their ability to dynamically download and install new software. Certainly, it may happen that an application is too demanding for a given underlying hardware revision. As these new devices are built in a monolithic way, a hardware upgrade is not an option. Nonetheless, most users do not want to buy a new device every time this happens. Thus, it is necessary to provide a possibility, which allows the processor to adapt to a given application at runtime, and thereby improving its own performance. This thesis presents three major approaches to such a runtime dynamic application acceleration.:1 Introduction 5 1.1 Motivation 5 1.2 Targets and Aims 7 1.3 Thesis Outline 8 2 AMIDAR - A Runtime Reconfigurable Processor 11 2.1 Overall Processor Architecture 11 2.2 Principle of Operation 14 2.3 Applicability of the AMIDAR Model 15 2.4 Adaptivity in AMIDAR Processors 16 2.5 Relations to Existing Processor Architectures 19 3 Applicability to Different Instruction Set Architectures 23 3.1 Supported Instruction Set Architectures 23 3.2 Selecting an ISA for Hardware Acceleration 25 3.3 A Detailed Look at an AMIDAR Based Java Processor 29 3.4 Example Token Sequence and Execution Trace 31 3.5 Performance Comparison of AMIDAR and IA32 Processors 34 4 Hotspot Evaluation 37 5 Runtime Reconfiguration of Processors 41 5.1 The Idea of Processor Reconfiguration 41 5.2 Targets and Aims for Efficient Processor Extensibility 43 6 Hardware Synthesis 47 6.1 The Evolution of Coarse Grain Reconfigurable Computing 47 6.2 The CGRA Target Architecture 71 6.3 Hardware Synthesis 79 6.4 Evaluation and Results of Hardware Synthesis 97 6.5 Saving Hardware With Heterogeneous CGRAs 103 6.6 The Size of Token Sets for Synthesized Functional Units 107 6.7 The Runtime Consumption of Performance Acceleration 108 7 Instruction Folding 113 7.1 The General Idea Behind Instruction Folding 113 7.2 General Classification of Folding Strategies 114 7.3 Folding Based on Instruction Type Pattern 116 7.4 Java Bytecode Folding Based on Behavioural Pattern 121 7.5 Common Applications of Instruction Folding 125 7.6 Instruction Folding and the AMIDAR Execution Model 126 8 Assembly of Microinstruction Groups 151 8.1 Motivation and General Idea 151 8.2 The Basic Token Set Assembly Algorithm 159 8.3 Algorithmic Extensions 179 8.4 Synthilation for an Unaltered Basic Processor 182 8.5 Synthilation Performance on Multi-ALU Processors 191 8.6 Runtime Characteristics of Synthilation Algorithms 195 9 Comparison 197 9.1 Speedup Comparison 197 9.2 Runtime and Complexity 198 9.3 Token Memory Consumption 200 9.4 Consumed Hardware Resources 201 10 Conclusion 203 10.1 Realization of Targets and Aims 203 10.2 The Ideal Use Case for Each Acceleration Approach 204 10.3 Limitations and Drawbacks 206 10.4 Summary 207 A Benchmark Applications 209 A.1 Cryptographic Ciphers 209 A.2 Hash Functions and Message Digests 210 A.3 Image Processing Filters 212 A.4 Jpeg Encoder 212 B Benchmark Measurement Values 213 B.1 Measurements of Instruction Set Evaluation 213 B.2 Measurement Values of Hardware Synthesis 217 B.3 Measurement Values of Instruction Folding 227 B.4 Measurement Values of Token Set Synthilation 243

Es wird ein Ansatz zur zeitbeschränkten Ablaufplanung für VLIW-Prozessoren mit neuronalen Netzen vorgestellt. Bestehende Arbeiten werden dahingehend erweitert, dass der Datenpfad des Prozessors über heterogene funktionale Einheiten verfügen und geclustert sein darf. Es werden zwei Varianten zur Lösung des Problems angegeben, deren Qualität mit einem heuristischen Ansatz verglichen wird und Schlussfolgerungen bezüglich der Nutzbarkeit neuronaler Netze gezogen.