کنفرانس مهندسی برق ایران

صفحه اصلی / سی و دومین کنفرانس بین المللی مهندسی برق

Multi-Objective Concurrent Kernel Scheduling for Multi-GPU Systems

نویسندگان :

Negar Baradar Alizadeh¹ Mahmoud Momtazpour²

1- دانشگاه صنعتی امیرکبیر 2- دانشگاه صنعتی امیرکبیر

کلمات کلیدی :

GPU،Concurrent Kernel Execution،Scheduling،Energy Consumption،Quality of Service،Genetic Algorithm

چکیده :

GPUs are now being widely used in many applications to speed up the computations. Due to its unique architecture, GPUs are well suited to run embarrassingly parallel applications such as deep learning and big data analysis. However, GPUs are power hungry devices, and many applications are not able to fully utilize the resources of a GPU, resulting in energy wastage in multi-GPU systems. To tackle this challenge, recent advancement has made it possible to run multiple kernels concurrently on a single GPU to share its resources. This however comes with a performance penalty due to the interference of concurrent kernels while competing over shared resources. In the presence of such impacts, mapping and scheduling of multiple kernels on multiple GPUs in order to fully utilize the GPUs, minimize the energy consumption, and reduce interference and performance degradation is challenging. In this paper, we propose a multi-objective kernel scheduling approach for multi-GPU systems, which considers both energy consumption and performance and supports concurrent kernel execution. To solve the scheduling problem, due to its large solution space, a GA-based scheduler called CK-GA has been used that can achieve desirable solutions at a reasonable pace. We also propose a heuristic scheduling algorithm called CK-HEFT based on the well-known HEFT algorithm for concurrent kernel execution on shared GPUs. Experimental results on several multi-GPU architectures show that the proposed CK-GA scheduler has obtained more than 14% improvement in performance, 30% improvement in quality of service and more than 18% improvement in energy consumption on average compared to similar approaches.

لیست مقالات

لیست مقالات بایگانی شده

An Autonomous Multi Agent Q-Learning Approach for Resource Allocation in D2D-Enabled Heterogeneous Networks

Pouya Akhoundzadeh - Ghasem Mirjalily - Mohammad taghi Saadeghi

Extension Network of Radiomics-based Deeply Supervised U-Net (ERDU) For Prostate Image Segmentation

Mahdi Ashtarian - Karim Faez - Marjan Firouznia - Hamidreza Amindavar

تحلیل دینامیکی ماشین سنکرون مغناطیس دائم با آهنربای جانبی و تحلیل خطای اتصال کوتاه داخلی و ضعیف شدن آهنربا

آزیتا فتحی - پیمان نادری

Compare Efficiency of Modeled and Designed Tilt Bi-rotor, Quad-rotor, and Hexa-rotor in Agriculture Application

Arshia Rezaei - Amin Talaeizadeh - Aria Alasti

Physiotherapy Algorithms on FUM-Physio Robot

Keyvan Tayaranian Marvian - Amir Hossein Nazari - Seyed Mohammad Tahamipour Zarandi - Mohammad Reza Akbarzadeh totonchi - Zahra Soltani - Alireza Akbarzadeh totonchi

Adaptive fault tolerant neural control of heterogeneous second-order multi-agent systems

Mohammad Hadi Rezaei - Ali Abooee

Analysis of the DC Bias Effects on the Transformer Vibration Using a Multi-field Coupling Model

Amir Esmaeili Nezhad - Mohammad Hamed Samimi

Synergizing ISAC and OTFS in a Non-GB-OMA Downlink Framework

Ghasem Saeidi - Hamid Saeedi-sourck

Revealing Shadows: Low-Light Image Enhancement Using Self-Calibrated Illumination

Farzaneh Koohestani - Nader Karimi - Shadrokh Samavi

Improving the Performance of ST-GCN on Multi-Site rs-fMRI Data Through Time Repetition Alignment

Mehrana Calagari - Hamidreza Hakimdavoodi

بیشتر

ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 41.7.4