Tesseract A 4D Network Control Plane

1
TesseractA 4D Network Control Plane

• Carnegie Mellon University
• Microsoft Research
• Rice University
• Presented by Alberto Gonzalez, Whitney Young

2
Current Designs

• No direct control
• Subtle dependencies
• Example load balance forwarding by tuning OSPF
  link weights, but impacts inter-domain routing

3
4D Architecture

• Control plane
• Decision
• Dissemination
• Discovery
• Data

• Services
• Dissemination
• Node configuration

4
Design
5
Design Goals

• Timely reaction to network changes
• Resilient to decision plane failure
• Robust and secure control channels
• Minimal switch configuration
• Backward compatibility
• Support diverse decision algorithms
• Support multiple data planes

6
Implementation Overview

• Switch
• Implements data plane
• Decision Element (DE)
• Implements discovery, dissemination, and decision
  planes

7
Decision Plane

• Any network control algorithm can be easily
  integrated
• Incremental shortest path first
• Spanning tree
• Joint packet filtering/routing
• Link cost-based traffic engineering
• Resiliency to DE failure
• Hot standbys receiving heartbeats

8
Dissemination Plane

• Goal communication between DEs and switches
• DEs handle most of dissemination plane, but
  switches help out
• Path to destination handled by DE
• Switches have separate queue and dissemination
  packets have higher priority
• Security (protects switches, info passed through
  dissemination plane, and compromised DEs)

9
Discovery Plane

• Goal minimize manual configuration
• Switches send HELLO messages
• DEs handle instructing the switches on what to do
  once active
• Initiate eBGP session with outside world
• Backward compatibility (bootstrapping end hosts)
• Discovery plane as DHCP proxy

10
Data Plane

• Configured by decision plane
• WriteTable exposed with simple interface to
  provide configuration service to decision plane
• Allows easy implementation of different services
• Decision/Dissemination Interface
• Function independently of each other
• Only 3 functions used to interface between them
  (2 more simply to improve performance)

11
Performance Evaluation

• Single Link Failures

• Switch Regional Failures

• Link Flapping
• 10-hop to 12-hop change
• Tesseract can handle network changes

12
Performance Evaluation

• 1347 nodes 6244 edges
• DE Computation Time
• Worst Case 151ms
• 99th percentile 40ms
• Bandwidth overhead
• Worst Case 4.4MB
• 90 of switched updated with new state

13
Performance Evaluation

• Failover times

14
Applications

• In enterprise network
• Computers both new routes packet filter
  placements
• Loads into routers with no forbidden traffic
  leaked
• No human involvement once security policy is
  specified

15
Ethernet

• Key features
• Widely implemented frame format
• Support for broadcasting frames
• Transparent address learning model
• Tesseract keeps these properties.

16
Ethernet

• Through point comparisons
• Control Plane for TCP flows
• Started at 570Mbps
• Leveled at 280Mbps after a failure
• Conventional RSTP Control Plane
• Starts at 280Mbps
• Hit zero after failure
• Recovered after 7-8 seconds at 180Mbps

17
Summary

• Tesseract
• Robust
• Decission/Dissemination Planes
• Secure
• Enterprise Network
• Resuable
• Ethernet or IP
• Good Performance
• Convergence Throughput
• Scalable
• 1,000 Switches
• Enables direct Control
• Easier to Understand and Deploy