3. Background
Infrastructure Usage Utilization is not cost effective.
Total cost can be reduced by sharing network resources
Cloud Solutions (Cloud Computing, etc) do not take
infrastructure as concern.
Always take network connectivity for granted and do not
approach considering QoS [1]
Requirement of connection between components
respecting service requirements.
Virtualized service platform respecting all service
requirements, e.g. as expressed by interactive real-time
services, on transport layer [1]
4. Definition
”A promising approach to cover individual and dynamic
resource provision while keeping strong individual QoS
requirements and optimizing the overall resource
usage” , Oberle [1].
"A technique for isolating computational and network
resources through virtualization to allocate them to a
logical (virtual) network for accommodating multiple
independent and programmable virtual networks" ,
Nakao [2].
”A mechanism for running multiple networks, which are
customized to a specific purpose, over the shared
infrastructure” , Miyamura [3].
5. Key Features
Segmentation: allows several different services to share a
physical link with given specific QoS properties
Isolation: No crosstalk between applications in resource
sharing caused by program crashes, sniffing, attacking, etc
Encapsulation: enables services developers to design service
specific on the overlay networks at a high level of abstraction,
and then disburden them fromdealing with highly complex
physical network infrastructures.
Aggregation: possible to build virtually elarged resources
(clustered resources or resources pool)
7. ISONI Architecture
ISONI : Intellegent Service Oriented Network Infrastructure
ISONI characteristic:
Reduce complexity for roll-out new services
Separate management hardware resources from
services
Upper Part : VSN (Virtual Service Network) provided by
service developer
Lower Part : Real Resources, VMU (Virtual Machine Unit)
and link between VMU
11. CABO Architecture
CABO : Concurrent Architecture Better than One
Divide ISP into two distinct entities:
infrastructure providers (provide physically)
service providers (provide agrement)
Physical Resources shared by subdividing physical to
virtual for both node and links
Bandwidth and delay guaranteed by arbritate access to
shared resources (CPU, memory, and bandwidth)
End host can run multiple virtual network from different
service provider
13. Network Control Mechanism
One-Hop Source Routing
Based on Routing overlays
Route to intermediate node, relay to destination by
ordinary IP routing, forwarding by tunnel
Simple control and scalable
Adaptive network control mechanism
Attractor selection based VNT based on environmental
changes adapation
Regulatory and metabolic reactions consider as optical
and service overlay network
14. Resource Allocation
Static Approaches
Basic Algorithm
Traffic constraints based algorithm
Splitting and Migration of Paths
Dynamic Approaches
DaVinci
Miscellaneous Approaches
Autonomic Systems based
Control Theoretic based systems
16. VNET on HEN
Consist of 110 computers and single non-blocking gigabit
etherne switch with constant latency
Node and Link Virtualization Technologies for Instantion
Rely on XEN's Paravirtualization for hosting virtual
machines
Physical Node compose substrate (PIP)
NOC of VNP connect to dedicated management node
XML Schema describe resource spesification for Node and
Links
18. VNET on HEN
Two option node virtualization :
VM created and booted as guest domain
VM resources allocated by PIP upon request
Virtual node connection using Ipv4-in-IPv4 Tunnel
Two NIC drivers of XEN :
DomU (Back-end) : part kernel space of guest OS
Dom0 (Front-end): physical domain inc, physical NIC
Back-end and Front-end correspons with Bridging (XEN
default) or Click
Topology created using VLAN and Virtual links by switch-
daemon based on request
19. VINI on PlanetLab
PlanetLab : large physical infrastructure and provide
virtualization
Vserver of the node (Slice) for experiment isolation
Tight control of resources (CPU or Bandwidth per slice)
CPU scheduler ”fair share”, Linux Hierarchical Token
Bucket (HTB) ”fair share and minimum rate guarantees”
VNET module for track and multiplexes incoming and
outgoing traffic
UML (User Mode Linux) allow each virtual node access
multiple network device (user-space in a slice)
21. VINI on PlanetLab
Linux TAP/TUN driver modification to send and receive
packet on the overlay
Single TUN/TAP interface (same IP address) used by
multiple processes (different slices) simultaneously
IIAS (Internet In A Slice) : example network architecture for
evaluate existing routing and forwarding mechanism
IIAS employs Click Software Router (forwarding engine),
and XORP routing protocol suite (control plane)
XORP run in UML kernel process, FIB implemented in
Click Process outside UML
Next Development GpENI-VINI : MyPLC (VINI resources
manager) and IIAS (interface and link provisioning tools)
23. Reference
1) Karsten Oberle, Marcus Kessler, Manuel Stein, Thomas Voith,
Dominik Lamp, Sören Berger, "Network Virtualization: The
missing piece", ICIN, 2009.
2) Akihiro NAKAO, "Network Virtualization as Foundation for
Enabling New Network Architectures and Applications", IEICE,
March 2010.
3) Takashi Miyamura, Yuichi Ohsita, Shin’ichi Arakawa, Yuki
Koizumi, Akeo Masuda, Kohei Shiomoto, and Masayuki Murata,
"Network Virtualization Server for Adaptive Network Control",
ITC Specialist Seminar, Hanoi, 2009.
4) Panagiotis Papadimitriou, Olaf Maennel, Adam Greenhalgh,
Anja Feldmann, Laurent Mathy, ”Implementing Network
Virtualization for a Future Internet”, Hoi An, Vietnam, 2009.
24. Reference
5) Panagiotis Papadimitriou, Olaf Maennel, Adam Greenhalgh,
Anja Feldmann, Laurent Mathy, ”Network Virtualization
Architecture: Proposal and Initial Prototype", VISA, Spain, 2009.
6) Nick Feamster, Lixin Gao, Jennifer Rexford, "How to Lease the
Internet in Your Spare Time".
7) K. Tutschku, T. Zinner, A. Nakao, P. Tran-Gia, "Network
Virtualization: Implementation Steps Towards the Future
Internet", Electronic Communications of the EASST Volume 17,
2009.
8) Aun Haider, Richard Potter, Akihiro Nakao, "Challenges in
Resource Allocation in Network Virtualization", ITC Specialist
Seminar, Hanoi, 2009.
25. Reference
9) Ramkumar Cherukuri, Xuan Liu , Andy Bavier, James P.G.
Sterbenz, and Deep Medhi, "Network Virtualization in GpENI:
Framework, Implementation & Integration Experience",
IEEE/IFIP International Workshop, Ireland, 2011.
10) Andy Bavier, Nick Feamster, Mark Huang, Larry Peterson,
Jennifer Rexford, "In VINI Veritas: Realistic and Controlled
Network Experimentation".