Advanced Memory Optimization Techniques for Low-Power by Manish Verma, Peter Marwedel

By Manish Verma, Peter Marwedel

The layout of embedded structures warrants a brand new viewpoint as a result following purposes: to start with, gradual and effort inefficient reminiscence hierarchies have already turn into the bottleneck of the embedded structures. it's documented within the literature because the reminiscence wall challenge. Secondly, the software program working at the modern embedded units is changing into more and more complicated. it's also good understood that no silver bullet exists to unravel the reminiscence wall challenge. for this reason, this publication explores a collaborative strategy by way of presenting novel reminiscence hierarchies and software program optimization innovations for the optimum usage of those reminiscence hierarchies. Linking reminiscence structure layout with memory-architecture acutely aware compilation ends up in speedy, energy-efficient and timing predictable reminiscence accesses. The evaluate of the optimization options utilizing real-life benchmarks for a unmarried processor approach, a multiprocessor system-on-chip (SoC) and for a electronic sign processor procedure, studies major discount rates within the strength intake and function development of those structures. The ebook provides a variety of optimizations, gradually expanding within the complexity of study and of reminiscence hierarchies. the ultimate bankruptcy covers optimization thoughts for purposes such as a number of methods present in newest embedded units. complex reminiscence Optimization strategies for Low strength Embedded Processors is designed for researchers, complier writers and embedded process designers / architects who desire to optimize the power and function features of the reminiscence subsystem.

Show description

Read or Download Advanced Memory Optimization Techniques for Low-Power Embedded Processors PDF

Similar microprocessors & system design books

The Scientist and Engineers Guide to Digital Signal Processing

The publication provides the basics of DSP utilizing examples from universal technology and engineering challenge

Programming and Customizing the Avr Microcontroller

This reader-friendly advisor exhibits you ways to take cost of the latest, such a lot flexible microcontrollers round, Atmel's AVR RISC chip family members. within, Electronics international author and astronomy instrumentation developer Dhananjay V. Gadre walks you from first assembly those intriguing new computers-on-a-chip throughout layout and ready-to-launch items.

Modular low-power, high-speed CMOS analog-to-digital converter of embedded systems

One of many major traits of microelectronics is towards layout for built-in platforms, i. e. , system-on-a-chip (SoC) or system-on-silicon (SoS). because of this improvement, layout thoughts for mixed-signal circuits develop into extra vital than sooner than. between different units, analog-to-digital and digital-to-analog converters are the 2 bridges among the analog and the electronic worlds.

Model-Driven Design Using IEC 61499: A Synchronous Approach for Embedded and Automation Systems

This ebook describes a singular technique for the layout of embedded structures and business automation structures, utilizing a unified model-driven strategy that's appropriate in either domain names. The authors illustrate their method, utilizing the IEC 61499 normal because the major car for specification, verification, static timing research and automatic code synthesis.

Extra info for Advanced Memory Optimization Techniques for Low-Power Embedded Processors

Sample text

2 Compilation Framework The compilation framework for a uni-processor ARM is based on the energy optimizing C compiler ENCC [37]. 4, ENCC takes application source code written in ANSI C [7] as input and generates an optimized assembly file containing Thumb mode instructions. The assembly file is then assembled and linked using the standard tool chain from ARM, and the executable binary of the application is generated. 1 Uni-Processor ARM 23 specific intermediate representation also known as IR-C.

Total execution cycles, memory accesses, energy consumption values for processors and memories are collected. Fig. 7. Multi-Process Edge Detection Application Multi-Process Edge Detection Benchmark: The memory optimizations for the multi-processor ARM based system are evaluated for the multi-process edge detection benchmark. The original benchmark was obtained from [50] and was parallelized so that it can execute on a multi-processor system. The multi-processor benchmark consists of an initiator process, a terminator process and a variable number of compute processes to detect the edges in the input tomographic images.

In contrast, the SA approach achieves energy efficient allocations for all the other scratchpad sizes. The other important observation is that the energy values for the Frac. SA are always very close to those for the SA approach. A maximum difference of 2% is observed for the dsp benchmark at 512 bytes scratchpad sizes. Therefore, if the system architecture permits allocation of a memory object across the boundary of a scratchpad, then the Frac. SA approach having a polynomial time complexity should be used to replace the SA approach.

Download PDF sample

Rated 4.46 of 5 – based on 47 votes