Integrated System-Level Modeling of Network-on-Chip enabled Multi-Processor PlatformsSpringer Science & Business Media, 2006. gada 25. aug. - 186 lappuses We are presently observing a paradigm change in designing complex SoC as it occurs roughly every twelve years due to the exponentially increasing number of transistors on a chip. This design discontinuity, as all previous ones, is characterized by a move to a higher level of abstraction. This is required to cope with the rapidly increasing design costs. While the present paradigm change shares the move to a higher level of abstraction with all previous ones, there exists also a key difference. For the ?rst time shrinking geometries do not leadtoacorrespondingincreaseofperformance. InarecenttalkLisaSuofIBM pointed out that in 65nm technology only about 25% of performance increase can be attributed to scaling geometries while the lion share is due to innovative processor architecture [1]. We believe that this fact will revolutionize the entire semiconductor industry. What is the reason for the end of the traditional view of Moore’s law? It is instructive to look at the major drivers of the semiconductor industry: wireless communications and multimedia. Both areas are characterized by a rapidly increasingdemandofcomputationalpowerinordertoprocessthesophisticated algorithmsnecessarytooptimallyutilizethepreciousresourcebandwidth. The computational power cannot be provided by traditional processor architectures and shared bus type of interconnects. The simple reason for this fact is energy ef?ciency: there exist orders of magnitude between the energy ef?ciency of an algorithm implemented as a ?xed functionality computational element and of a software implementation on a processor. |
No grāmatas satura
1.–5. rezultāts no 39.
xi. lappuse
... efficiency: there exist orders of magnitude between the energy efficiency of an algorithm implemented as a fixed functionality computational element and of a software implementation on a processor. We argue that future SoC for wireless ...
... efficiency: there exist orders of magnitude between the energy efficiency of an algorithm implemented as a fixed functionality computational element and of a software implementation on a processor. We argue that future SoC for wireless ...
xii. lappuse
... efficient execution of the functional task. The second, and equally important, task is concerned with the inter-task data exchanges which have to be mapped onto an interconnect architecture. Both computation and communication have seen ...
... efficient execution of the functional task. The second, and equally important, task is concerned with the inter-task data exchanges which have to be mapped onto an interconnect architecture. Both computation and communication have seen ...
1. lappuse
... efficiency [3]. The growing potential for silicon integration is even outpaced by the amount of functionality incorporated into embedded devices from all kinds of applica- tion domains. This trend originates from the tremendous increase ...
... efficiency [3]. The growing potential for silicon integration is even outpaced by the amount of functionality incorporated into embedded devices from all kinds of applica- tion domains. This trend originates from the tremendous increase ...
2. lappuse
... efficient execution of the functional tasks. Additionally, the inter-task data exchange has to be mapped to a commu- nication architecture. Both processing and communication mapping are highly interrelated and only a joint consideration ...
... efficient execution of the functional tasks. Additionally, the inter-task data exchange has to be mapped to a commu- nication architecture. Both processing and communication mapping are highly interrelated and only a joint consideration ...
3. lappuse
... efficiency by tailoring instruction set and micro architecture to the respective set of tasks [11]. Examples are innovative architectures exploiting Instruction Level Parallelism (ILP) as well as Data Level Parallelism (DLP) [12] ...
... efficiency by tailoring instruction set and micro architecture to the respective set of tasks [11]. Examples are innovative architectures exploiting Instruction Level Parallelism (ILP) as well as Data Level Parallelism (DLP) [12] ...
Saturs
9 | |
CLASSIFICATION OF PLATFORM ELEMENTS | 15 |
SYSTEM LEVEL DESIGN PRINCIPLES | 33 |
RELATED WORK | 43 |
METHODOLOGY OVERVIEW | 59 |
UNIFIED TIMING MODEL | 79 |
MPSOC SIMULATION FRAMEWORK | 113 |
CASE STUDY | 141 |
SUMMARY | 153 |
B The OCPIP TL3 Channel | 163 |
List of Figures | 171 |
List of Tables | 175 |
About the Authors | 195 |
Citi izdevumi - Skatīt visu
Integrated System-Level Modeling of Network-on-Chip enabled Multi-Processor ... Tim Kogel,Rainer Leupers,Heinrich Meyr Priekšskatījums nav pieejams - 2006 |
Integrated System-Level Modeling of Network-on-Chip enabled Multi-Processor ... Tim Kogel,Rainer Leupers,Heinrich Meyr Priekšskatījums nav pieejams - 2010 |
Integrated System-Level Modeling of Network-on-Chip enabled Multi-Processor ... Tim Kogel,Rainer Leupers,Heinrich Meyr Priekšskatījums nav pieejams - 2009 |
Bieži izmantoti vārdi un frāzes
Abstract Data Types abstraction levels algorithm application domain application model arbitration architecture models Automation Conference DAC bandwidth Based Design blocks bus architectures CEFSM Co-design communication architecture Communication Node complex configuration CoWare crossbar cycle-level TLM data-plane processing defined denotes depicted in figure Design Automation Conference design space exploration DiffServ efficiency embedded applications Embedded Systems enables engine execution FIFO Functional Process hardware heterogeneous HW/SW IEEE implementation Initiator input signals IPv4 ISO/OSI reference model latency layer metrics module MP-SoC platform multi-processor multi-threaded Network Processor Network-on-Chip NoC framework on-chip communication on-chip networks OSCI output signal packet packet switch packet-level paradigm parameters performance Proc processing delay processing elements processor protocol queuing Reactive Process Network requirements RWTH Aachen simulation speed Software switches synchronization interface Synthesis system architect System Level Design System-on-Chip SystemC target throughput TL3 API topology Virtual Architecture Mapping VPU Node