Performance characterization of illumination algorithms for reconfigurable graphics processor

来源 :中国邮电高校学报(英文版) | 被引量 : 0次 | 上传用户:h459403474
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
Graphics processing is an increasing important application domain with the demand of real-time rendering,video streaming,virtual reality,and so on.Illumination is a critical module in graphics rendering and is typically compute-bound,memory-bound,and power-bound in different application cases.It is crucial to decide how to schedule different illumination algorithms with different features according to the practical requirements in reconfigurable graphics hardware.This paper analyze the performance characteristics of four main-stream lighting algorithms,Lambert illumination algorithm,Phong illumination algorithm,Blinn-Phong illumination algorithm,and Cook-Torrance illumination algorithm,using hardware performance counters on x86 processor platform KabyLake (KBL).The data movement,computation,power consumption,and memory accessing are evaluated over a range of application scenarios.Further,by analyzing the system-level behavior of these illumination algorithms,obtains the cons and pros of these specific algorithms were obtained.The associated relationship between performance/energy and the evaluated metrics was analyzed through Pearson correlation coefficient (PCC) analysis.According to these performance characterization data,this paper presents some reconfiguration suggestions in reconfigurable graphics processor.
其他文献
The indoor positioning system based on fingerprint receives more and more attention due to its high positioning accuracy and time efficiency.In the existing positioning approaches,much consideration is given to the positioning accuracy improvement by usin
The independent hypothesis between frames in vocal effect (VE) recognition makes it difficult for frame based spectral features to describe the intrinsic temporal correlation and dynamic change information in speech phenomena.A novel VE detection method b
A 12-bit 2.6 GS/s radio frequency digital to analog converter (RF DAC) based on 1 μm GaAs heterojunction bipolar transistor (HBT) process is presented.The DAC integrates a 4∶1 multiplexer to reduce the data rate of input ports,which greatly facilitates th
A system model consisting of macro and micro base stations (BS) is introduced to solve the problem of power allocation in heterogeneous dense network.In this hierarchical framework,the problem of power allocation is modeled as a stackelberg game.Based on
A novel single-cavity equilateral triangular substrate integrated waveguide (TSIW) bandpass filter (BPF) with a complementary triangular split ring resonator (CTSRR) is designed in this paper.A metallic via-hole is used to split the degenerate modes and a
Achterbahn-128 is a stream cipher proposed by Gammel et al.and submitted to the eSTREAM project.Though many attacks have been published,no recovery attack better than Naya-Plasencia\'s results with 256 bit keystream limitation.Similar approach is shown
High efficiency video coding (HEVC) uses half of the bitrate compared to H.264/advanced video coding (AVC) for encoding the same sequence with similar quality.Because of the advanced hierarchical structures of coding units (CUs),predicting units (PUs),and
With the development of wireless network and electronic technologies,the wireless sensor network (WSN) has been widely used in many applications.One of the most important applications is wireless medical sensor network (WMSN),which makes modern health-car
Flocking of multi-agent system with high-frequency feedback robust control is put forward in detail.The controller that has high-frequency feedback robust control is designed;it is proved that the velocity error is bounded and no collision occurs between
Hoh Xil is the national nature reserve in China,and Tibetan antelope is a research hotspot of wildlife protection in this area.In order to track the population and activity of Tibetan antelope in Hoh Xil,a quantum wireless sensor monitoring network (QWSMN