IaaS Providers need next-generation virtualized networks, providing large-scale multi-tenancy services and meeting new requirements in terms of efficiency and availability. The main limitation of current network architectures mainly comes from inadequate design standards. In this presentation, we present a TRILL based multi-tenant network architecture for virtualized data center networks. This architecture allows the best of both layers (bridging and routing), thus building a large-scale layer-2 network while ensuring scalability, efficiency, fault-tolerance and simplified management. This enables also the creation of large numbers of logical networks within a virtualized data center. The kernel part of our solution is currently developped in the Linux kernel.

1. VIRTUAL NETWORK OVER
TRILL
DESIGN, IMPLEMENTATION AND DEMONSTRATION
- Gandi.netWilliam Dauchy
Kernel Recipes 2013

2. CONVENTIONAL DATA CENTER

3. MAIN GOAL
provide large scale multi-tenancy

4. LARGE SCALE MULTI-TENANCY
Multiples users using same resources

5. REQUIREMENTS
Seamless VM mobility
Easy management
Layer 2 core scaling
Fault resiliance
VLAN scalability

6. LAYER 2 - SWITCHING BENEFITS
Management simplified + Plug & play
Seamless Virtual Machine mobility
Auto learning + determistic failover

7. LAYER 2 - SWITCHING LIMITATION
A large number of tenants implies
a huge number of MAC address in switch table (TCAM
overflow)
ARP storm at nodes
STP to ensure a loop free topology
blocking redundant paths
Core-computes required, recomputes when topology
changes
Number of VLANs is limited to 4096

9. TRILL BASED DATA CENTER

10. WHAT IS TRILL
New device: RBridge
Control plane
Data plane
Encapsulate native frames in a transport header
Providing a hop count and nickname
Route the encapsulated frames using IS-IS
Decapsulate native frames before delivery

11. IETF STANDARD
RFC 5556 Transparent Interconnection of Lots of Links (TRILL): Problemand Applicability Statement
RFC 6325 Routing Bridges (RBridges): Base Protocol Specification
RFC 6326 Transparent Interconnection of Lots of Links (TRILL) Use of IS-IS
RFC 6327 Routing Bridges (RBridges): Adjacency
RFC 6439 Routing Bridges (RBridges): Appointed Forwarders
RFC 6361 PPP Transparent Interconnection of Lots of Links (TRILL) Protocol Control Protocol

12. DESIGN AND IMPLEMENTATION

13. CONTROL PLANE
unicast building

14. CONTROL PLANE
unicast building - first iteration

15. CONTROL PLANE
unicast building - second iteration

16. CONTROL PLANE
unicast building - third iteration

17. CONTROL PLANE
unicast building - final result

18. CONTROL PLANE

19. CONTROL PLANE
multicast building - first iteration

20. CONTROL PLANE
multicast building - final iteration

21. DATA PLANE

22. DATA PLANE

23. DATA PLANE

24. DATA PLANE

25. DATA PLANE

26. IMPLEMENTATION - SENDING

27. IMPLEMENTATION - RECEIVING

28. LAYER 2 - SWITCHING LIMITATION
A large number of tenants implies
a huge number of MAC address in switch table
ARP storm at nodes
STP to ensure a loop free topology
blocking redundant paths
Core-computes required, recomputes when topology
changes
Number of VLANs is limited to 4096

29. TRILL + VNI = VNT
Virtual Network over TRILL

30. VNT FRAME FORMAT

31. VNI LIFE

32. VNI TOPOLOGY BUILDING

33. VNI TOPOLOGY BUILDING

34. VNI TOPOLOGY BUILDING

35. IMPLEMENTATION WITH VNI - SENDING

36. IMPLEMENTATION WITH VNI - RECEIVING

37. LINUX BIG PICTURE

38. DEMONSTRATION

39. SCREENCAST
screencast
(live explanation to understand what's going on)

40. PH.D. STUDY
Ahmed Amamou - ahmed@gandi.net
"Network isolation for Virtualized Datacenters"
University Pierre & Marie Curie - GANDISAS
project still in development and cleaning
TRILL sources: github.com/Gandi/ktrill
VNT: still two research projects working on it - drafts

41. GANDI.NET
Gandi Hosting - gandi.net/hosting
William Dauchy - william@gandi.net
slides pres.gandi.net/kr2013