Proceedings of the International Conference on Hardware/Software Codesign and System Synthesis Companion 2019
DOI: 10.1145/3349567.3351719
|View full text |Cite
|
Sign up to set email alerts
|

A simulation framework for domain-specific system-on-chips

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2020
2020
2020
2020

Publication Types

Select...
2
1

Relationship

2
1

Authors

Journals

citations
Cited by 3 publications
(4 citation statements)
references
References 9 publications
0
4
0
Order By: Relevance
“…This section provides an extension to our previous work in [41], using built-in DS3 schedulers and applications in the benchmark suite. The simulations run on an SoC configuration that mimics a typical heterogeneous SoC with a total of 16 general purpose cores and hardware accelerators: 4 big Arm Cortex-A15 cores, 4 LITTLE Arm Cortex-A7, 2 scrambler accelerators, 4 FFT accelerators, and 2 Viterbi decoders.…”
Section: Scheduler Case Studiesmentioning
confidence: 99%
“…This section provides an extension to our previous work in [41], using built-in DS3 schedulers and applications in the benchmark suite. The simulations run on an SoC configuration that mimics a typical heterogeneous SoC with a total of 16 general purpose cores and hardware accelerators: 4 big Arm Cortex-A15 cores, 4 LITTLE Arm Cortex-A7, 2 scrambler accelerators, 4 FFT accelerators, and 2 Viterbi decoders.…”
Section: Scheduler Case Studiesmentioning
confidence: 99%
“…The WiFi transmitter and receiver applications process 64 bits of data in one frame and are segmented into the kernels shown in Figure 7. It is composed of various compute-intensive blocks, such as FFT, modulation, demodulation, Viterbi decoder, and scrambler [13]. Range detection and Pulse Doppler applications are used in Radar to determine the distance and velocity, respectively, of the target object from the reference signal source.…”
Section: B Hardware Platforms and Applicationsmentioning
confidence: 99%
“…Our framework is successfully able to expose the limitations of underlying design decisions related to the SoC configuration and scheduling policies for a given set of applications. Traditionally, researchers use discrete event-based simulation tools, such as DS3 [13] and SimGrid [14], to develop and evaluate new scheduling algorithms. These simulators rely on statistical profiling information to realize the performance of general-purpose cores and hardware accelerators.…”
Section: Case Study 2: Performance Modementioning
confidence: 99%
See 1 more Smart Citation