MétaCan
Menu
Back to cohort
Record W4391216112 · doi:10.1109/tccn.2024.3358565

Priority-Aware Deployment of Autoscaling Service Function Chains Based on Deep Reinforcement Learning

2024· article· en· W4391216112 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueIEEE Transactions on Cognitive Communications and Networking · 2024
Typearticle
Languageen
FieldComputer Science
TopicSoftware-Defined Networks and 5G
Canadian institutionsYork University
FundersNational Key Research and Development Program of ChinaNational Natural Science Foundation of China
KeywordsComputer scienceSoftware deploymentQuality of serviceReinforcement learningScheduling (production processes)Computer networkDistributed computingCloud computingArtificial intelligenceOperating systemEngineering

Abstract

fetched live from OpenAlex

Communication networks are being restructured by means of network function virtualization (NFV) and service-based architecture (SBA) to embrace greater flexibility, agility, programmability and efficiency. The deployment of service function chains (SFCs) to flexibly offer diverse network services is considered essential in NFV-based networks. Beyond the fifth-generation (5G) and sixth-generation (6G) eras, SFC deployment should be capable of satisfying various quality of service (QoS) requirements, coping with dynamic network states and traffic, handling urgent business in a timely manner, and avoiding resource congestion, all of which present significant scheduling challenges. In this paper, we propose a priority-aware deployment framework for autoscaling and multi-objective SFCs, which mainly includes 2 parts. First, to guarantee the diverse QoS requirements (e.g., latency and request acceptance rate) of various network services, a multi-objective SFC deployment scheme is established to optimize the service latency, deployment cost and service acceptance rate. Second, a deep reinforcement learning (DRL) algorithm, named the autoscaling and priority-aware SFC deployment algorithm (APSD), is further designed to solve the multi-objective optimization problem, which is NP hard. In APSD, we first prioritize requests with varying real-time characteristics to ensure that urgent services can be processed in a timely manner; based on the resiliency characteristics of virtual network functions (VNFs), we propose a hybrid scaling strategy to scale VNFs both horizontally and vertically to respond to changes in service requests and workload. We report comprehensive experiments carried out to assess the effectiveness of the proposed SFC deployment framework and demonstrate its advantages over its counterparts. Thus, we show that APSD is time efficient in solving the multi-objective optimization problem and that the obtained strategy always consumes the least resources (e.g., central processing unit (CPU) and memory resources) and surpasses two baseline algorithms with a 29.5% and 12.36% lower latency on average.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.994
Threshold uncertainty score0.850

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.033
GPT teacher head0.271
Teacher spread0.238 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it