Traffic Classification for Application Specific Peering

1
Traffic Classification for Application Specific
Peering

• Jia Wang
• Joint work with Balachander Krishnamurthy
• ATT Labs Research
• November 7, 2002

2
Motivation

• Application specific peering
• P2P-like protocols
• Allow searches for resources to be directed to a
  copy of a resource on an ISPs network
• Traffic classification graph transformation
• AS relationship
• Flow size

3
Background

• Inter-AS relationship
• Provider-customer customer pays provider
• Peer-peer mutually benefit by exchanging traffic
  between respective customers
• BGP export rules
• An AS can export its routes and routes to its
  customers to its provides/peers, but can not
  export routes learnt from other providers/peers.
• An AS can exports its routes, routes of its
  customers and routes learnt from other
  providers/peers to its customers.

4
Methodology

• AS paths

• AS categories
• ISP
• ISP-CUST
• ISP-PEER
• ISP-CUST-CUST
• ISP-PEER-CUST
• ISP-MH-CUST
• UNKNOWN

provider
provider
customer
customer
provider
provider
customer
customer
peer
provider
provider
customer
peer
customer
provider
provider
customer
customer
5
Traffic classification
P2P traces
Netflow
AS relationship
BGP configuration
P2P traffic flows
ISP ISP-PEER ISP-CUST
ISP-CUST-CUST ISP-PEER-CUST ISP-MH-CUST
Graph representation
Connected Components
Threshold
Signaling traffic Data traffic
Traffic classification
6
Traffic data

• P2P applications DirectConnent, FastTrack,
  Gnutella
• Flow records from routers across a large IP
  network
• Control traffic flow size lt threshold
• Data traffic flow size gt threshold
• Threshold 4KB
• Experiments 3 weeks, each 5-7 days, 800 million
  flows in total from ATT IP backbone

7
Classification results

• Gnutella heaviest connected component
• 99 IPs, 99 bytes
• Signaling traffic 90 IPs, 95 flows, 0.4 bytes
• Data traffic 50 IPs, 5 flows, 99.6 bytes
• Signaling traffic is much less skewed than data
  traffic
• Signaling top 1 IPs, 25 bytes
• Data top 0.1 IPs, 30 bytes

8
Traffic direction
Type Signaling Signaling Data Data
Direction Src () Dst () Src () Dst ()
ATT 1.00 3.77 0.72 3.08
ATT-CUST 18.46 62.05 20.85 60.75
ATT-PEER 10.58 6.54 9.29 6.86
ATT-CUST-CUST 0.00 0.00 0.00 0.00
ATT-PEER-CUST 59.38 12.05 59.18 14.73
ATT-MH-CUST 4.99 8.22 5.65 7.13
5 of the traffic are intra-ATT (ATT and
ATT-CUST).
9
Conclusion

• Traffic classification
• AS relationship and flow size
• Graph transformation
• Protocol-independent
• Scalable
• Automatic
• Applications
• Feasibility of retaining the proper portion of
  traffic on an ISPs network to reduce the costs
  and latency
• Overylay topology efficiency