{"id":{"repo_id":"ets-quebec","oai_identifier":"oai:espace.etsmtl.ca:3191"},"canonical_url":"https://search.ndltd.org/etd/ets-quebec/oai:espace.etsmtl.ca:3191","repository":{"repo_id":"ets-quebec","name":"ETS (Quebec)","base_url":"https://espace.etsmtl.ca/cgi/oai2"},"display":{"title":"LSTM-Based VNF resource usage forecasting leveraging SFC resource attribute interdependencies","abstract":"As of late, Network Function Virtualization (NFV) has emerged as a research topic garnering a lot of interest, thanks to its many purported benefits to Next-Generation Network (NGN) infrastructures such as 5G, 6G and IoT. Intense research is currently being conducted to design robust algorithms that can accurately predict resource usage of various entities in NFV infrastructures (NFVIs), such as Virtualized Network Functions (VNFs), Service Function Chains (SFCs), server nodes and server clusters, with the aim to scale or migrate those VNFs or SFCs properly. To date, however, few research teams have focused their efforts on finding useful dependencies and relationships that these entities might have with each other. For example, scaling a specific VNF might impact the CPU resource usage and network traffic of other VNFs in the same SFC. In another case, migrating a whole SFC from one server cluster to another might negatively impact the network traffic of some network slices dedicated to control or encryption functions of that SFC and other similar SFC types. To tackle the issue of identifying VNF dependencies, Deep Learning (DL) and, more particularly, Recurrent Neural Networks (RNNs) might prove extremely useful due to their proven ability to decipher deep, hidden relationships between multiple features in time series. Hence, the aim of this research project will be to find means of identifying VNF and SFC dependencies in NFVIs and cloud computing environments and to leverage VNF and SFC dependency models to enhance dynamic VNF resource usage forecasting methods in multi-VNF environments. To do so, our goal will be 1) to identify and classify different VNF dependencies, 2) build ETL (Extract, Transform, Load) mechanisms to adapt sequential VNF resource usage history from multiple features for DL, and finally, 3) design a VNF resource usage forecasting mechanism leveraging resource attribute interdependencies in an SFC.","abstract_html":"As of late, Network Function Virtualization (NFV) has emerged as a research topic garnering a lot of interest, thanks to its many purported benefits to Next-Generation Network (NGN) infrastructures such as 5G, 6G and IoT. Intense research is currently being conducted to design robust algorithms that can accurately predict resource usage of various entities in NFV infrastructures (NFVIs), such as Virtualized Network Functions (VNFs), Service Function Chains (SFCs), server nodes and server clusters, with the aim to scale or migrate those VNFs or SFCs properly. To date, however, few research teams have focused their efforts on finding useful dependencies and relationships that these entities might have with each other. For example, scaling a specific VNF might impact the CPU resource usage and network traffic of other VNFs in the same SFC. In another case, migrating a whole SFC from one server cluster to another might negatively impact the network traffic of some network slices dedicated to control or encryption functions of that SFC and other similar SFC types. To tackle the issue of identifying VNF dependencies, Deep Learning (DL) and, more particularly, Recurrent Neural Networks (RNNs) might prove extremely useful due to their proven ability to decipher deep, hidden relationships between multiple features in time series. Hence, the aim of this research project will be to find means of identifying VNF and SFC dependencies in NFVIs and cloud computing environments and to leverage VNF and SFC dependency models to enhance dynamic VNF resource usage forecasting methods in multi-VNF environments. To do so, our goal will be 1) to identify and classify different VNF dependencies, 2) build ETL (Extract, Transform, Load) mechanisms to adapt sequential VNF resource usage history from multiple features for DL, and finally, 3) design a VNF resource usage forecasting mechanism leveraging resource attribute interdependencies in an SFC.","abstract_has_math":false,"creators":["St-Onge, Cédric"],"institution":"École de technologie supérieure","degree_name":"Thèse de doctorat électronique","degree_level":"phd","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-02-03","date_published":"2023-02-03","updated_at":"2026-11-18T22:09:45Z","subjects":["adaptation de ressources VNF, prédiction d’utilisation de ressources VNF, interdépendances VNF, LSTM, coefficients de corrélation, réseaux neuronal en graphe, filtrage de valeurs aberrantes multivariées"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3191"],"render_values":[{"text":"INSERT-YOUR-LIBRARY-ID-HERE3191","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null,"record_url":null,"record_label":null,"record_kind":null,"record_source":null,"download_url":"https://espace.etsmtl.ca/id/eprint/3191/1/ST_ONGE_Cedric.pdf","download_label":"Download PDF","download_source":"dc:identifier","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://espace.etsmtl.ca/cgi/oai2?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aespace.etsmtl.ca%3A3191","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["St-Onge, Cédric"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2023-02-03"]},{"key":"dc:publisher","label":"Institution","values":["École de technologie supérieure"]},{"key":"dc:type","label":"Dc Type","values":["Electronic Thesis or Dissertation"]},{"key":"thesis:degree_level","label":"Degree Level","values":["phd"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Thèse de doctorat électronique"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["École de technologie supérieure"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["adaptation de ressources VNF, prédiction d’utilisation de ressources VNF, interdépendances VNF, LSTM, coefficients de corrélation, réseaux neuronal en graphe, filtrage de valeurs aberrantes multivariées"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3191","https://espace.etsmtl.ca/id/eprint/3191/1/ST_ONGE_Cedric.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["As of late, Network Function Virtualization (NFV) has emerged as a research topic garnering a lot of interest, thanks to its many purported benefits to Next-Generation Network (NGN) infrastructures such as 5G, 6G and IoT. Intense research is currently being conducted to design robust algorithms that can accurately predict resource usage of various entities in NFV infrastructures (NFVIs), such as Virtualized Network Functions (VNFs), Service Function Chains (SFCs), server nodes and server clusters, with the aim to scale or migrate those VNFs or SFCs properly. To date, however, few research teams have focused their efforts on finding useful dependencies and relationships that these entities might have with each other. For example, scaling a specific VNF might impact the CPU resource usage and network traffic of other VNFs in the same SFC. In another case, migrating a whole SFC from one server cluster to another might negatively impact the network traffic of some network slices dedicated to control or encryption functions of that SFC and other similar SFC types. To tackle the issue of identifying VNF dependencies, Deep Learning (DL) and, more particularly, Recurrent Neural Networks (RNNs) might prove extremely useful due to their proven ability to decipher deep, hidden relationships between multiple features in time series. Hence, the aim of this research project will be to find means of identifying VNF and SFC dependencies in NFVIs and cloud computing environments and to leverage VNF and SFC dependency models to enhance dynamic VNF resource usage forecasting methods in multi-VNF environments. To do so, our goal will be 1) to identify and classify different VNF dependencies, 2) build ETL (Extract, Transform, Load) mechanisms to adapt sequential VNF resource usage history from multiple features for DL, and finally, 3) design a VNF resource usage forecasting mechanism leveraging resource attribute interdependencies in an SFC."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["LSTM-Based VNF resource usage forecasting leveraging SFC resource attribute interdependencies"]}]}],"canonical_facts":{"dc:creator":["St-Onge, Cédric"],"dc:date":["2023-02-03"],"dc:description":["As of late, Network Function Virtualization (NFV) has emerged as a research topic garnering a lot of interest, thanks to its many purported benefits to Next-Generation Network (NGN) infrastructures such as 5G, 6G and IoT. Intense research is currently being conducted to design robust algorithms that can accurately predict resource usage of various entities in NFV infrastructures (NFVIs), such as Virtualized Network Functions (VNFs), Service Function Chains (SFCs), server nodes and server clusters, with the aim to scale or migrate those VNFs or SFCs properly. To date, however, few research teams have focused their efforts on finding useful dependencies and relationships that these entities might have with each other. For example, scaling a specific VNF might impact the CPU resource usage and network traffic of other VNFs in the same SFC. In another case, migrating a whole SFC from one server cluster to another might negatively impact the network traffic of some network slices dedicated to control or encryption functions of that SFC and other similar SFC types. To tackle the issue of identifying VNF dependencies, Deep Learning (DL) and, more particularly, Recurrent Neural Networks (RNNs) might prove extremely useful due to their proven ability to decipher deep, hidden relationships between multiple features in time series. Hence, the aim of this research project will be to find means of identifying VNF and SFC dependencies in NFVIs and cloud computing environments and to leverage VNF and SFC dependency models to enhance dynamic VNF resource usage forecasting methods in multi-VNF environments. To do so, our goal will be 1) to identify and classify different VNF dependencies, 2) build ETL (Extract, Transform, Load) mechanisms to adapt sequential VNF resource usage history from multiple features for DL, and finally, 3) design a VNF resource usage forecasting mechanism leveraging resource attribute interdependencies in an SFC."],"dc:format":["application/pdf"],"dc:identifier":["INSERT-YOUR-LIBRARY-ID-HERE3191","https://espace.etsmtl.ca/id/eprint/3191/1/ST_ONGE_Cedric.pdf"],"dc:language":["en"],"dc:publisher":["École de technologie supérieure"],"dc:subject":["adaptation de ressources VNF, prédiction d’utilisation de ressources VNF, interdépendances VNF, LSTM, coefficients de corrélation, réseaux neuronal en graphe, filtrage de valeurs aberrantes multivariées"],"dc:title":["LSTM-Based VNF resource usage forecasting leveraging SFC resource attribute interdependencies"],"dc:type":["Electronic Thesis or Dissertation"],"thesis:degree_level":["phd"],"thesis:degree_name":["Thèse de doctorat électronique"],"thesis:institution_name":["École de technologie supérieure"]},"updated_at":"2026-11-18T22:09:45Z"}