Difference between revisions of "Rashid Mijumbi"

From SimpleWiki
Jump to navigationJump to search
 
(One intermediate revision by the same user not shown)
Line 39: Line 39:
 
Network virtualisation is a promising technique for dealing with the resistance of the Internet to architectural changes,  enabling a novel business model in which infrastructure management is decoupled from service provision. It allows infrastructure providers (InPs) who own substrate networks (SNs) to lease chunks of them out to service providers who then create virtual networks (VNs), which can then be re-leased out or used to provide services to end-users.
 
Network virtualisation is a promising technique for dealing with the resistance of the Internet to architectural changes,  enabling a novel business model in which infrastructure management is decoupled from service provision. It allows infrastructure providers (InPs) who own substrate networks (SNs) to lease chunks of them out to service providers who then create virtual networks (VNs), which can then be re-leased out or used to provide services to end-users.
  
However,  the different VNs should be initialised, in which case virtual links and nodes must be mapped to substrate nodes and paths respectively. One of the challenges in the initialisation of VNs is the requirement of an efficient sharing of SN resources. Since the profitability of InPs depends on how many VNs are able to be allocated simultaneously onto the SN, the success of network virtualisation will depend, in part, on how efficiently VNs utilise physical network resources. This thesis contributes to efficient resource sharing in network virtualisation by dividing the problem into three sub-problems: (1) mapping virtual nodes and links to substrate nodes and paths i.e. virtual network embedding (VNE), (2) dynamic managing of the resources allocated to VNs throughout their lifetime (DRA), and (3) provisioning of backup resources to ensure survivability of the VNs.
+
However,  the different VNs should be initialised, in which case virtual links and nodes must be mapped to substrate nodes and paths respectively. One of the challenges in the initialisation of VNs is the requirement of an efficient sharing of SN resources. Since the profitability of InPs depends on how many VNs are able to be allocated simultaneously onto the SN, the success of network virtualisation will depend, in part, on how efficiently VNs utilise physical network resources. This PhD Research contributes to efficient resource sharing in network virtualisation by dividing the problem into three sub-problems: (1) mapping virtual nodes and links to substrate nodes and paths i.e. virtual network embedding (VNE), (2) dynamic managing of the resources allocated to VNs throughout their lifetime (DRA), and (3) provisioning of backup resources to ensure survivability of the VNs.
  
The constrained VNE problem is NP-Hard. As a result, to simplify the solution, many existing approaches propose heuristics that make assumptions (e.g. a SN with infinite resources), some of which would not apply in practical environments. This thesis proposes an improvement in VNE by proposing a one-shot VNE algorithm which is based on column generation (CG). The CG approach starts by solving a restricted version of the problem, and thereafter refines it to obtain a final solution. The objective of a one-shot mapping is to achieve better resource utilisation, while using CG significantly enhances the solution time complexity.
+
The constrained VNE problem is NP-Hard. As a result, to simplify the solution, many existing approaches propose heuristics that make assumptions (e.g. a SN with infinite resources), some of which would not apply in practical environments. This research proposes an improvement in VNE by proposing a one-shot VNE algorithm which is based on column generation (CG). The CG approach starts by solving a restricted version of the problem, and thereafter refines it to obtain a final solution. The objective of a one-shot mapping is to achieve better resource utilisation, while using CG significantly enhances the solution time complexity.
  
In addition current approaches are static in the sense that after the VNE stage, the resources allocated are not altered for the entire lifetime of the VN. The few proposals that do allow for adjustments in original mappings allocate a fixed amount of node and link resources to VNs throughout their life time. Since network load varies with time due to changing user demands, allocating a fixed amount of resources based on peak load could lead to an inefficient utilisation of overall SN resources, whereby, during periods when some virtual nodes and/or links are lightly loaded, SN resources are still reserved for them, while possibly rejecting new VN requests. The second contribution of this thesis are a set of proposals that ensure that SN resources are efficiently utilised, while at the same making sure that the QoS requirements of VNs are met. For this purpose, we propose self-management algorithms in which the SN uses time-difference machine learning techniques to make autonomous decisions with respect to resource allocation.
+
In addition current approaches are static in the sense that after the VNE stage, the resources allocated are not altered for the entire lifetime of the VN. The few proposals that do allow for adjustments in original mappings allocate a fixed amount of node and link resources to VNs throughout their life time. Since network load varies with time due to changing user demands, allocating a fixed amount of resources based on peak load could lead to an inefficient utilisation of overall SN resources, whereby, during periods when some virtual nodes and/or links are lightly loaded, SN resources are still reserved for them, while possibly rejecting new VN requests. The second contribution of this research are a set of proposals that ensure that SN resources are efficiently utilised, while at the same making sure that the QoS requirements of VNs are met. For this purpose, we propose self-management algorithms in which the SN uses time-difference machine learning techniques to make autonomous decisions with respect to resource allocation.
  
Finally, while some scientific research has already studied multi-domain VNE, the available approaches to survivable VNs have focused on the single InP environment. Since in the more practical situation a network virtualisation environment will involve multiple InPs, and because an extension of network survivability approaches from the single to multi domain environments is not trivial, this thesis proposes a distributed and dynamic approach to survivability in VNs. This is achieved by using a multi-agent-system that uses a multi-attribute negotiation protocol and a dynamic pricing model forming InPs coalitions supporting  SNs resource backups. The ultimate objective is to ensure that virtual network operators maximise  profitability by minimising penalties resulting from QoS violations.
+
Finally, while some scientific research has already studied multi-domain VNE, the available approaches to survivable VNs have focused on the single InP environment. Since in the more practical situation a network virtualisation environment will involve multiple InPs, and because an extension of network survivability approaches from the single to multi domain environments is not trivial, this PhD research proposes a distributed and dynamic approach to survivability in VNs. This is achieved by using a multi-agent-system that uses a multi-attribute negotiation protocol and a dynamic pricing model forming InPs coalitions supporting  SNs resource backups. The ultimate objective is to ensure that virtual network operators maximise  profitability by minimising penalties resulting from QoS violations.
  
 
== Recent publications ==
 
== Recent publications ==
 +
*Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Meng Shen, Ke Xu, Kun Yang, A Neuro-Fuzzy Approach to Self-Management of Virtual Network Resources,  Journal of Expert Systems With Applications. Accepted August 2014.
 
* Rashid Mijumbi, Joan Serrat and Juan Luis Gorricho, Learning Algorithms for Dynamic Resource Allocation in Virtualised Networks, Management of Large Scale Virtualized Infrastructures: Smart Data Acquisition, Analysis and Network and Service Management in the Future Internet.  June 2014.
 
* Rashid Mijumbi, Joan Serrat and Juan Luis Gorricho, Learning Algorithms for Dynamic Resource Allocation in Virtualised Networks, Management of Large Scale Virtualized Infrastructures: Smart Data Acquisition, Analysis and Network and Service Management in the Future Internet.  June 2014.
 
* Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Maxim Claeys, Jeroen Famaey, Filip De Turck, Neural Network-based Autonomous Allocation of Resources in Virtual Networks, European Conference on Networks and Communications (EuCNC), 2014, Bologna, Italy. Accepted, April 2014.
 
* Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Maxim Claeys, Jeroen Famaey, Filip De Turck, Neural Network-based Autonomous Allocation of Resources in Virtual Networks, European Conference on Networks and Communications (EuCNC), 2014, Bologna, Italy. Accepted, April 2014.

Latest revision as of 20:58, 24 August 2014

Summary
Student: Rashid Mijumbi
Title: Self-managed Resources in Network Virtualisation Environments
e-mail: rashid@tsc.upc.edu
Affiliation: Universitat Politecnica de Catalunya
Supervisor: Prof. Joan Serrat and Dr. Juan-Luis Gorricho
Start: October 2010
End: October 2014
Funding: Government of Spain

Biography

Rashid graduated from Makerere University (Kampala, Uganda) with a Bachelors Degree in Electrical Engineering. He is currently a PhD Candidate in the department of Signal Theory and Communications at the Technical University of Catalonia (Barcelona, Spain); where he is also a member in the Research group Management, Pricing and Services in Next Generation Networks (MAPS).

Research interests

His research interests are in the management of networks and services for the future Internet. Current focus is on resource management in network virtualisation environments. Medium term focus will be on SDN-based network virtualisation and Network Function Virtualisation (NFV).

PhD Project Description

Network virtualisation is a promising technique for dealing with the resistance of the Internet to architectural changes, enabling a novel business model in which infrastructure management is decoupled from service provision. It allows infrastructure providers (InPs) who own substrate networks (SNs) to lease chunks of them out to service providers who then create virtual networks (VNs), which can then be re-leased out or used to provide services to end-users.

However, the different VNs should be initialised, in which case virtual links and nodes must be mapped to substrate nodes and paths respectively. One of the challenges in the initialisation of VNs is the requirement of an efficient sharing of SN resources. Since the profitability of InPs depends on how many VNs are able to be allocated simultaneously onto the SN, the success of network virtualisation will depend, in part, on how efficiently VNs utilise physical network resources. This PhD Research contributes to efficient resource sharing in network virtualisation by dividing the problem into three sub-problems: (1) mapping virtual nodes and links to substrate nodes and paths i.e. virtual network embedding (VNE), (2) dynamic managing of the resources allocated to VNs throughout their lifetime (DRA), and (3) provisioning of backup resources to ensure survivability of the VNs.

The constrained VNE problem is NP-Hard. As a result, to simplify the solution, many existing approaches propose heuristics that make assumptions (e.g. a SN with infinite resources), some of which would not apply in practical environments. This research proposes an improvement in VNE by proposing a one-shot VNE algorithm which is based on column generation (CG). The CG approach starts by solving a restricted version of the problem, and thereafter refines it to obtain a final solution. The objective of a one-shot mapping is to achieve better resource utilisation, while using CG significantly enhances the solution time complexity.

In addition current approaches are static in the sense that after the VNE stage, the resources allocated are not altered for the entire lifetime of the VN. The few proposals that do allow for adjustments in original mappings allocate a fixed amount of node and link resources to VNs throughout their life time. Since network load varies with time due to changing user demands, allocating a fixed amount of resources based on peak load could lead to an inefficient utilisation of overall SN resources, whereby, during periods when some virtual nodes and/or links are lightly loaded, SN resources are still reserved for them, while possibly rejecting new VN requests. The second contribution of this research are a set of proposals that ensure that SN resources are efficiently utilised, while at the same making sure that the QoS requirements of VNs are met. For this purpose, we propose self-management algorithms in which the SN uses time-difference machine learning techniques to make autonomous decisions with respect to resource allocation.

Finally, while some scientific research has already studied multi-domain VNE, the available approaches to survivable VNs have focused on the single InP environment. Since in the more practical situation a network virtualisation environment will involve multiple InPs, and because an extension of network survivability approaches from the single to multi domain environments is not trivial, this PhD research proposes a distributed and dynamic approach to survivability in VNs. This is achieved by using a multi-agent-system that uses a multi-attribute negotiation protocol and a dynamic pricing model forming InPs coalitions supporting SNs resource backups. The ultimate objective is to ensure that virtual network operators maximise profitability by minimising penalties resulting from QoS violations.

Recent publications

  • Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Meng Shen, Ke Xu, Kun Yang, A Neuro-Fuzzy Approach to Self-Management of Virtual Network Resources, Journal of Expert Systems With Applications. Accepted August 2014.
  • Rashid Mijumbi, Joan Serrat and Juan Luis Gorricho, Learning Algorithms for Dynamic Resource Allocation in Virtualised Networks, Management of Large Scale Virtualized Infrastructures: Smart Data Acquisition, Analysis and Network and Service Management in the Future Internet. June 2014.
  • Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Maxim Claeys, Jeroen Famaey, Filip De Turck, Neural Network-based Autonomous Allocation of Resources in Virtual Networks, European Conference on Networks and Communications (EuCNC), 2014, Bologna, Italy. Accepted, April 2014.
  • Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Contributions to Efficient Resource Management in Virtual Networks, 8th International Conference on Autonomous Infrastructure, Management and Security (AIMS), June 30 - July 3, 2014, Brno, Czech Republic. Accepted, March 2014.
  • Rashid Mijumbi, Juan-Luis Gorricho, Joan Serrat, Maxim Claeys, Steven Latre, Filip De Turck, Design and Evaluation of Learning Algorithms for Dynamic Resource Management in Virtual Networks, IEEE/IFIP Networks Operations and Management Symposium, (NOMS), 2014. Best Student Paper Award.
  • Rashid Mijumbi, Joan Serrat and Juan Luis Gorricho, Multi-Agent Based Resource Allocation in Next Generation Virtual Networks, Proceedings of the 2012 Barcelona Forum on Ph.D. Research in Communication and Information Technologies. October, 2012.
  • Rashid Mijumbi, Joan Serrat and Juan Luis Gorricho, Autonomic Resource Management in Virtual Network, Scalable and Adaptive Internet Solutions (SAIL) Summer School Work in Progress Session, Santander, Cantabria. June 2012. Best Work in Progress Award.

Invited Talks

  • Learning Algorithms for Dynamic Resource Allocation in Virtualised Networks, Workshop on Management of Large Scale Virtualized Infrastructures: Smart Data Acquisition, Analysis and Network and Service Management in the Future Internet, co-located with European Conference on Networks and Communications (EuCNC), Bologna, Italy. June 2014. (Invited).
  • Application of Learning Techniques to Resource Management in Virtual Networks, Future Internet Technologies, Tsinghua University, Beijing, PR China. December 2013.

Research Visits

  • March 22, 2014 - May 02, 2014. Information Technology Laboratory, Center for Research and Advanced Studies of the National Polytechnic Institute of Mexico (CINVESTAV), 87130 Ciudad Victoria, Tamaulipas, Mexico.
  • October 02, 2013 - January 25, 2014. Tsinghua University, 100084, Beijing, PR China.
  • May 6, 2013 - May 18, 2013. Ghent University 􀀀 iMinds, B-9050 Ghent, Belgium.

External links