Acceleration of radiance for lighting simulation by using parallel computing with OpenCL

Wangda Zuo; Andrew Mcneil; Michael Wetter; Eleanor Lee

Acceleration of radiance for lighting simulation by using parallel computing with OpenCL

Wangda Zuo, Andrew Mcneil, Michael Wetter, Eleanor Lee

Architectural Engineering

Research output: Contribution to conference › Paper › peer-review

5 Scopus citations

Abstract

We report on the acceleration of annual daylighting simulations for fenestration systems in the Radiance ray-tracing program. The algorithm was optimized to reduce both the redundant data input/output operations and the floating-point operations. To further accelerate the simulation speed, the calculation for matrix multiplications was implemented using parallel computing on a graphics processing unit. We used OpenCL, which is a cross- platform parallel programming language. Numerical experiments show that the combination of the above measures can speed up the annual daylighting simulations 101.7 times or 28.6 times when the sky vector has 146 or 2306 elements, respectively.

Original language	English (US)
Pages	110-117
Number of pages	8
State	Published - 2011
Event	12th Conference of International Building Performance Simulation Association Building Simulation 2011, BS 2011 - Sydney, NSW, Australia Duration: Nov 14 2011 → Nov 16 2011

Other

Other	12th Conference of International Building Performance Simulation Association Building Simulation 2011, BS 2011
Country/Territory	Australia
City	Sydney, NSW
Period	11/14/11 → 11/16/11

All Science Journal Classification (ASJC) codes

Computer Science Applications
Building and Construction
Architecture
Modeling and Simulation

Cite this

@conference{997a12e4b3bf4668a78c81842b7d96bf,

title = "Acceleration of radiance for lighting simulation by using parallel computing with OpenCL",

abstract = "We report on the acceleration of annual daylighting simulations for fenestration systems in the Radiance ray-tracing program. The algorithm was optimized to reduce both the redundant data input/output operations and the floating-point operations. To further accelerate the simulation speed, the calculation for matrix multiplications was implemented using parallel computing on a graphics processing unit. We used OpenCL, which is a cross- platform parallel programming language. Numerical experiments show that the combination of the above measures can speed up the annual daylighting simulations 101.7 times or 28.6 times when the sky vector has 146 or 2306 elements, respectively.",

author = "Wangda Zuo and Andrew Mcneil and Michael Wetter and Eleanor Lee",

year = "2011",

language = "English (US)",

pages = "110--117",

note = "12th Conference of International Building Performance Simulation Association Building Simulation 2011, BS 2011 ; Conference date: 14-11-2011 Through 16-11-2011",

}

Acceleration of radiance for lighting simulation by using parallel computing with OpenCL. / Zuo, Wangda; Mcneil, Andrew; Wetter, Michael et al.
2011. 110-117 Paper presented at 12th Conference of International Building Performance Simulation Association Building Simulation 2011, BS 2011, Sydney, NSW, Australia.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Acceleration of radiance for lighting simulation by using parallel computing with OpenCL

AU - Zuo, Wangda

AU - Mcneil, Andrew

AU - Wetter, Michael

AU - Lee, Eleanor

PY - 2011

Y1 - 2011

N2 - We report on the acceleration of annual daylighting simulations for fenestration systems in the Radiance ray-tracing program. The algorithm was optimized to reduce both the redundant data input/output operations and the floating-point operations. To further accelerate the simulation speed, the calculation for matrix multiplications was implemented using parallel computing on a graphics processing unit. We used OpenCL, which is a cross- platform parallel programming language. Numerical experiments show that the combination of the above measures can speed up the annual daylighting simulations 101.7 times or 28.6 times when the sky vector has 146 or 2306 elements, respectively.

AB - We report on the acceleration of annual daylighting simulations for fenestration systems in the Radiance ray-tracing program. The algorithm was optimized to reduce both the redundant data input/output operations and the floating-point operations. To further accelerate the simulation speed, the calculation for matrix multiplications was implemented using parallel computing on a graphics processing unit. We used OpenCL, which is a cross- platform parallel programming language. Numerical experiments show that the combination of the above measures can speed up the annual daylighting simulations 101.7 times or 28.6 times when the sky vector has 146 or 2306 elements, respectively.

UR - http://www.scopus.com/inward/record.url?scp=84870168693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870168693&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84870168693

SP - 110

EP - 117

T2 - 12th Conference of International Building Performance Simulation Association Building Simulation 2011, BS 2011

Y2 - 14 November 2011 through 16 November 2011

ER -

Acceleration of radiance for lighting simulation by using parallel computing with OpenCL

Abstract

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this