Is it fair? Resource allocation for differentiated services on demands

Zhang, Ran; Liu, Ning; Liu, Lei; Zhang, Wei; Yuan, Hao; Dong, Mianxiong; Cui, Lizhen

doi:10.1109/icws55610.2022.00059

Cited by 2 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Through these models, IoT nodes can upload information to a nearby centralized gateway by reusing the communication channels of traditional cellular users, thus achieving the allocation of underlying IoT resources. To obtain accurate performance prediction for resource allocation on demand, Zhang et al [12] designed a knowledge-driven multi-queue GPS performance prediction method, called DLPE, which manually selects relevant features and introduces them into the original features by analyzing the relevant theoretical knowledge of multi-queue GPS. In [12], the authors were the first to try to combine feature engineering traditional machine learning with a deep neural network.…”

Section: Deep Learning-based Methodsmentioning

confidence: 99%

“…Based on this, Zhang et al [11] gave an approximate analytical solution of the queue length distribution for multi-queue GPS. In order to further improve the accuracy of the multi-queue GPS performance prediction results, Zhang et al [12] designed a knowledge-driven performance prediction method based on a deep learning network by combining GPS theoretical analysis and deep learning.…”

Section: Approximate Analytical Methodsmentioning

confidence: 99%

“…Due to the limitations of analytical methods, only approximate solutions for performance prediction can be obtained. To obtain accurate performance metrics, Zhang et al [12] predicted the performance of the GPS scheduling mechanism by employing a knowledgedriven deep learning-based method, where queue theory and the attention network are combined. This marked a successful endeavor in leveraging deep learning-based methods for GPS performance prediction, thereby highlighting the considerable potential in this domain.…”

Section: Performance Prediction Of the Multi-queue Gps System In The ...mentioning

confidence: 99%

“…In [11], the authors used empty buffer approximate (EBA)-based analytical methods to derive the queue length distribution of each application service. In addition, to obtain accurate performance metrics, Zhang et al [12] tried to predict the performance of the GPS scheduling mechanism by employing a deep learning-based method, where the queue theory and the attention network are combined. Although there has been much work on the performance prediction for the GPS scheduling mechanism, limitations remain.…”

Section: Introductionmentioning

confidence: 99%

“…The performance prediction of the multi-queue GPS system is to evaluate the performance of each single queue, such as average queue length LP = {l p 1 , l p 2 , ..., l p M } and average queue delay LD = {ld 1 , ld 2 , ..., ld M }. Referring to the statement in [12], this problem is denoted as the multi-queue GPS problem.…”

mentioning

confidence: 99%

See 4 more Smart Citations

On-Demand Centralized Resource Allocation for IoT Applications: AI-Enabled Benchmark

Zhang,

Liu,

Dong

et al. 2024

Sensors

Self Cite

View full text Add to dashboard Cite

The development of emerging information technologies, such as the Internet of Things (IoT), edge computing, and blockchain, has triggered a significant increase in IoT application services and data volume. Ensuring satisfactory service quality for diverse IoT application services based on limited network resources has become an urgent issue. Generalized processor sharing (GPS), functioning as a central resource scheduling mechanism guiding differentiated services, stands as a key technology for implementing on-demand resource allocation. The performance prediction of GPS is a crucial step that aims to capture the actual allocated resources using various queue metrics. Some methods (mainly analytical methods) have attempted to establish upper and lower bounds or approximate solutions. Recently, artificial intelligence (AI) methods, such as deep learning, have been designed to assess performance under self-similar traffic. However, the proposed methods in the literature have been developed for specific traffic scenarios with predefined constraints, thus limiting their real-world applicability. Furthermore, the absence of a benchmark in the literature leads to an unfair performance prediction comparison. To address the drawbacks in the literature, an AI-enabled performance benchmark with comprehensive traffic-oriented experiments showcasing the performance of existing methods is presented. Specifically, three types of methods are employed: traditional approximate analytical methods, traditional machine learning-based methods, and deep learning-based methods. Following that, various traffic flows with different settings are collected, and intricate experimental analyses at both the feature and method levels under different traffic conditions are conducted. Finally, insights from the experimental analysis that may be beneficial for the future performance prediction of GPS are derived.

show abstract

Section: Deep Learning-based Methodsmentioning

confidence: 99%

Section: Approximate Analytical Methodsmentioning

confidence: 99%

Section: Performance Prediction Of the Multi-queue Gps System In The ...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations