Full-System Simulation of Mobile CPU/GPU Platforms

Kaszyk, Kuba; Wagstaff, Harry; Spink, Tom; Franke, Björn; O'Boyle, Mike; Bodin, Bruno; Uhrenholt, Henrik

doi:10.1109/ispass.2019.00015

Cited by 3 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prune=63 Prune=127 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.1x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x 1.0x for different sizes of a convolutional layer, as well as lowerlevel details about the execution in hardware, we executed the workloads in a Full-System Mali GPU simulator [22].…”

Section: B Channel Pruning Observed Through Gpu Simulationmentioning

confidence: 99%

“…Through the use of higher level libraries, like the ACL, we lose observability that we would normally have when working directly with OpenCL. To understand all the calls and kernel management performed by the Arm Compute Library for different sizes of a convolutional layer, as well as lowerlevel details about the execution in hardware, we executed the workloads in a Full-System Mali GPU simulator [22].…”

Section: B Channel Pruning Observed Through Gpu Simulationmentioning

confidence: 99%

See 1 more Smart Citation

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs

Radu

Kaszyk

Wen

et al. 2019

2019 IEEE International Symposium on Workload Characterization (IISWC)

Self Cite

View full text Add to dashboard Cite

Convolutional Neural Networks (CNN) are becoming a common presence in many applications and services, due to their superior recognition accuracy. They are increasingly being used on mobile devices, many times just by porting large models designed for server space, although several model compression techniques have been considered. One model compression technique intended to reduce computations is channel pruning. Mobile and embedded systems now have GPUs which are ideal for the parallel computations of neural networks and for their lower energy cost per operation. Specialized libraries perform these neural network computations through highly optimized routines. As we find in our experiments, these libraries are optimized for the most common network shapes, making uninstructed channel pruning inefficient. We evaluate higher level libraries, which analyze the input characteristics of a convolutional layer, based on which they produce optimized OpenCL (Arm Compute Library and TVM) and CUDA (cuDNN) code. However, in reality, these characteristics and subsequent choices intended for optimization can have the opposite effect. We show that a reduction in the number of convolutional channels, pruning 12% of the initial size, is in some cases detrimental to performance, leading to 2× slowdown. On the other hand, we also find examples where performance-aware pruning achieves the intended results, with performance speedups of 3× with cuDNN and above 10× with Arm Compute Library and TVM. Our findings expose the need for hardware-instructed neural network pruning.

show abstract

Section: B Channel Pruning Observed Through Gpu Simulationmentioning

confidence: 99%

Section: B Channel Pruning Observed Through Gpu Simulationmentioning

confidence: 99%

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs

Radu

Kaszyk

Wen

et al. 2019

2019 IEEE International Symposium on Workload Characterization (IISWC)

Self Cite

View full text Add to dashboard Cite

show abstract

Horus: A Modular GPU Emulator Framework

Elhelw

Pai

2020

2020 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)

View full text Add to dashboard Cite

Efficient Auto-Tuning of Parallel Programs with Interdependent Tuning Parameters via Auto-Tuning Framework (ATF)

Rasch

Schulze

Steuwer

et al. 2021

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Auto-tuning is a popular approach to program optimization: it automatically finds good configurations of a program’s so-called tuning parameters whose values are crucial for achieving high performance for a particular parallel architecture and characteristics of input/output data. We present three new contributions of the Auto-Tuning Framework (ATF), which enable a key advantage in general-purpose auto-tuning : efficiently optimizing programs whose tuning parameters have interdependencies among them. We make the following contributions to the three main phases of general-purpose auto-tuning: (1) ATF generates the search space of interdependent tuning parameters with high performance by efficiently exploiting parameter constraints; (2) ATF stores such search spaces efficiently in memory, based on a novel chain-of-trees search space structure; (3) ATF explores these search spaces faster, by employing a multi-dimensional search strategy on its chain-of-trees search space representation. Our experiments demonstrate that, compared to the state-of-the-art, general-purpose auto-tuning frameworks, ATF substantially improves generating, storing, and exploring the search space of interdependent tuning parameters, thereby enabling an efficient overall auto-tuning process for important applications from popular domains, including stencil computations, linear algebra routines, quantum chemistry computations, and data mining algorithms.

show abstract

Full-System Simulation of Mobile CPU/GPU Platforms

Cited by 3 publications

References 33 publications

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs

Horus: A Modular GPU Emulator Framework

Efficient Auto-Tuning of Parallel Programs with Interdependent Tuning Parameters via Auto-Tuning Framework (ATF)

Contact Info

Product

Resources

About