AOA: Adaptive Overclocking Algorithm on CPU-GPU Heterogeneous Platforms
نویسندگان
چکیده
Abstract Although GPUs have been used to accelerate various convolutional neural network algorithms with good performance, the demand for performance improvement is still continuously increasing. CPU/GPU overclocking technology brings opportunities further in CPU-GPU heterogeneous platforms. However, inevitably increases power of CPU/GPU, which not conducive energy conservation, efficiency optimization, or even system stability. How effectively constrain total remain roughly unchanged during a key issue designing adaptive algorithms. There are two factors solving this issue. Firstly, dynamic upper bound must be set reflect real-time behavior characteristics program so that algorithm can better meet unchanging constraints; secondly, instead independently at both CPU and GPU sides, coordinately on considered adapt load balance higher constraints. This paper proposes an Adaptive Overclocking Algorithm (AOA) platforms achieve goal while remains unchanged. AOA uses function $$F_k$$ F k describe variable introduces imbalance factor W realize coordinated overclocking. Through verification several types (Intel $$^\circledR $$ ® Xeon E5-2660 & NVIDIA Tesla K80; Intel Core™i9-10920X NIVIDIA GeForce RTX 2080Ti), achieves average 10.7% 4.4% savings. To verify effectiveness AOA, we compare other methods including automatic boost, highest static optimal
منابع مشابه
Heterogeneous Sparse Matrix Computations on Hybrid GPU/CPU Platforms
Hybrid GPU/CPU clusters are becoming very popular in the scientific computing community, as attested by the number of such systems present in the Top 500 list. In this paper, we address one of the key algorithms for scientific applications: the computation of sparse matrix-vector products that lies at the heart of iterative solvers for sparse linear systems. We detail how design patterns for sp...
متن کاملOptimization Techniques for Mapping Algorithms and Applications onto CUDA GPU Platforms and CPU-GPU Heterogeneous Platforms
Title of dissertation: OPTIMIZATION TECHNIQUES FOR MAPPING ALGORITHMS AND APPLICATIONS ONTO CUDA GPU PLATFORMS AND CPU-GPU HETEROGENEOUS PLATFORMS Jing Wu, Doctor of Philosophy, 2014 Dissertation directed by: Professor Joseph F JaJa, Department of Electrical and Computer Engineering An emerging trend in processor architecture seems to indicate the doubling of the number of cores per chip every ...
متن کاملAdaptive Partitioning for Irregular Applications on Heterogeneous CPU-GPU Chips
Commodity processors are comprised of several CPU cores and one integrated GPU. To fully exploit this type of architectures, one needs to automatically determine how to partition the workload between both devices. This is specially challenging for irregular workloads, where each iteration’s work is data dependent and shows control and memory divergence. In this paper, we present a novel adaptiv...
متن کاملSolving Sparse Differential Riccati Equations on Hybrid CPU-GPU Platforms
The numerical treatment of the linear-quadratic optimal control problem requires the solution of Riccati equations. In particular, the differential Riccati equations (DRE) is a key operation for the computation of the optimal control in the finite-time horizon case. In this work, we focus on large-scale problems governed by partial differential equations (PDEs) where, in order to apply a feedba...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Lecture Notes in Computer Science
سال: 2023
ISSN: ['1611-3349', '0302-9743']
DOI: https://doi.org/10.1007/978-3-031-22677-9_14