Hybrid functional- and instruction-level power modeling for embedded and heterogeneous processor architectures

Blume, Holger; Becker, Daniel; Rotenberg, L.; Botteck, Martin; Brakensiek, J.; Noll, Thomas

doi:10.1016/j.sysarc.2007.01.002

Cited by 24 publications

(8 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [62], [63] and [61], a functional level power model is used for estimating the power consumption of a C program without compiling it. It requires partitioning the target processor into functional units, and estimating some key parameters like the cache miss rate, external data memory access rate and the processing rate.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Decision Policies With Predictable Performance

et al. 2018

View full text Add to dashboard Cite

As methods and tools for Cyber-Physical Systems grow in capabilities and use, one-size-fits-all solutions start to show their limitations. In particular, tools and languages for programming an algorithm or modeling a CPS that are specific to the application domain are typically more usable, and yield better performance, than general-purpose languages and tools. In the domain of cardiac arrhythmia monitoring, a small, implantable medical device continuously monitors the patient's cardiac rhythm and delivers electrical therapy when needed. The algorithms executed by these devices are streaming algorithms, so they are best programmed in a streaming language that allows the programmer to reason about the incoming data stream as the basic object, rather than force her to think about lower-level details like state maintenance and minimization. Because these devices are resource-constrained, it is useful if the programming language allowed predictable performance in terms of processing runtime and energy consumption, or more general costs. StreamQRE is a declarative streaming programming language, with an efficient and portable implementation and strong theoretical guarantees. In particular, its evaluation algorithm guarantees constant cost (runtime, memory, energy) per data item, and also calculates upper bounds on the per-item cost. Such an estimate of the cost allows early exploration of the algorithmic possibilities, while maintaining a handle on worst-case performance, on the basis of which hardware can be designed and algorithms can be tuned. Abstract-As methods and tools for Cyber-Physical Systems grow in capabilities and use, one-size-fits-all solutions start to show their limitations. In particular, tools and languages for programming an algorithm or modeling a CPS that are specific to the application domain are typically more usable, and yield better performance, than general-purpose languages and tools. In the domain of cardiac arrhythmia monitoring, a small, implantable medical device continuously monitors the patient's cardiac rhythm and delivers electrical therapy when needed. The algorithms executed by these devices are streaming algorithms, so they are best programmed in a streaming language that allows the programmer to reason about the incoming data stream as the basic object, rather than force her to think about lower-level details like state maintenance and minimization. Because these devices are resource-constrained, it is useful if the programming language allowed predictable performance in terms of processing runtime and energy consumption, or more general costs. StreamQRE is a declarative streaming programming language, with an efficient and portable implementation and strong theoretical guarantees. In particular, its evaluation algorithm guarantees constant cost (runtime, memory, energy) per data item, and also calculates upper bounds on the per-item cost. Such an estimate of the cost allows early exploration of the algorithmic possibilities, while maintaining a handle on worst-case per...

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Decision Policies With Predictable Performance

et al. 2018

View full text Add to dashboard Cite

show abstract

“…1(b), model-based profiling [11][12][13][14][15][16][17] has a pre-defined instructions/functions table associated with individual CPU power consumption. Table content usually changes with hardware, platform and OS specifications.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Application developers use these results to estimate and optimize the energy consumption of their programs. Model-based profiling can be categorized as low-level [11,12] and high-level models [13][14][15][16][17][18]. Low-level models are associated with power consumption patterns derived from the system architecture level or processor instruction level, and evaluates CPU power consumption by simulation or through the use of virtualization software.…”

Section: Literature Reviewmentioning

confidence: 99%

“…It stores function invocations in a database and computes the average power cost for each function. Blume et al [15] proposed a hybrid power model for embedded systems at the function and architecture levels. The model records the executions of both processor instructions and functions to increase the accuracy of power consumption profiling and prediction.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Cloud-based power estimation and power-aware scheduling for embedded systems

Chen

Chiang

2015

Computers & Electrical Engineering

View full text Add to dashboard Cite

“…For these measurements, real processors or lower-level simulators are used together with a set of benchmark programs (scenarios) that stress the different functional blocks in the processor. In [3], a hybrid FLPA and instruction-level microprocessor power estimation approach is proposed. They have shown that by incorporating instruction dependent behavior in the FLPA models, the accuracy of the models can be further improved.…”

Section: Related Workmentioning

confidence: 99%

A High-level Microprocessor Power Modeling Technique Based on Event Signatures

Stralen

Pimentel

2008

J Sign Process Syst

View full text Add to dashboard Cite

This paper presents a technique for high-level power estimation of microprocessors. The technique, which is based on abstract execution profiles called 'event signatures', operates at a higher level of abstraction than commonly-used instruction-set simulator (ISS) based power estimation methods and should thus be capable of achieving good evaluation performance. As a consequence, the technique can be very useful in the context of early system-level design space exploration. In this paper, we also compare our power estimation results to those from the instruction-level simulators Wattch and Sim-Panalyzer. In these experiments, we demonstrate that with a good underlying power model, the signature-based power modeling technique can yield accurate estimations (a mean error of 3.1% compared to Wattch in our experiments). At the same time, our signature-based power modeling technique is at least an order of magnitude faster than the simulations performed by Wattch or Sim-Panalyzer.

show abstract

Hybrid functional- and instruction-level power modeling for embedded and heterogeneous processor architectures

Cited by 24 publications

References 4 publications

Real-Time Decision Policies With Predictable Performance

Real-Time Decision Policies With Predictable Performance

Cloud-based power estimation and power-aware scheduling for embedded systems

A High-level Microprocessor Power Modeling Technique Based on Event Signatures

Contact Info

Product

Resources

About