Evaluation of the Proximity between Web Clients and their Local DNS Servers

1
Evaluation of the Proximity between Web Clients
and their Local DNS Servers

• Z. Morley Mao
• UC Berkeley (zmao_at_eecs.berkeley.edu)
• C. Cranor, M. Rabinovich, O. Spatscheck, and J.
  Wang
• ATT Labs-Research
• F. Douglis
• IBM Research

2
Motivation

• Content Distribution Networks (CDNs)
• Attempt to deliver content from servers close to
  users

Internet
Cache server
Cache server
Cache server
3
DNS based server selection

• Originator problem
• Assumes that clients are close to their local DNS
  servers

Authoritative DNS server ns.service.com
Local DNS Server ns.myisp.net
A.GTLD-SERVERS.NET
Verify the assumption that clients are close to
their local DNS servers
4
Measurement setup

• Three components
• 1x1 pixel embedded transparent GIF image
• ltimg srchttp//xxx.rd.example.com/tr.gif
  height1 width1gt
• A specialized authoritative DNS server
• Allows hostnames to be wild-carded
• An HTTP redirector
• Always responds with 302 Moved Temporarily
• Redirect to a URL with client IP address embedded

1x1 transparent GIF
5
Embedded image request sequence
6
Measurement Data
7
Measurement statistics
8
Proximity metrics

• AS clustering
• Network clustering
• Traceroute divergence
• Roundtrip time correlation

9
AS clustering

• Autonomous System (AS)
• A single administrative entity with unified
  routing policy
• Observes if client and LDNS belong to the same AS

10
Network clustering

• Krishnamurthy,Wang sigcomm00
• Based on BGP routing information using the
  longest prefix match
• Each prefix identifies a network cluster
• Observes if client and LDNS belong to the same
  network cluster

11
Traceroute divergence
Probe machine
a

• Shaikh et al. infocom00
• Use the last point of
• divergence
• Traceroute divergence
• Max(3,4)4

b
1
1
2
2
3
3
4
12
Roundtrip time correlation

• Correlation between message roundtrip times from
  a probe site to the client and its LDNS server
• The probe site represents a potential cache
  server location
• A crude metric, highly dependent on the probe site

13
Aggregate statistics of AS/network clustering

• More than 13,000 ASes
• Close to 75 total ASes
• 440,000 unique prefixes
• Close to 25 of all possible network clusters
• ? We have a representative data set

14
Proximity analysisAS, network clustering

• AS clustering coarse-grained
• Network clustering fine-grained
• Most clients not in the same routing entity as
  their LDNS
• Clients with LDNS in the same cluster slightly
  more active

15
Proximity analysisTraceroute divergence

• Probe sites
• NJ(UUNET), NJ(ATT), Berkeley(Calren),
  Columbus(Calren)
• Sampled from top half of busy network clusters
• Median divergence 4
• Mean divergence 5.8-6.2
• Ratio of common to disjoint path length
• 72-80 pairs traced have common path at least as
  long as disjoint path

16
Improved local DNS configuration

• For client-LDNS associations not in the same
  cluster, do we know a LDNS in the clients
  cluster?

Client IPs
HTTP requests
17
Impact on commercial CDNs
Client w/ CDN server in cluster

• Data set
• Client-LDNS associations
• LDNS-CDN associations
• Available CDN servers

18
Impact on commercial CDNsAS clustering
19
Impact on commercial CDNsNetwork clustering
20
Conclusion

• Novel technique for finding client and local DNS
  associations
• Fast, non-intrusive, and accurate
• DNS based server selection works well for
  coarse-grained load-balancing
• 64 associations in the same AS
• 16 associations in the same network cluster
• Server selection can be inaccurate if server
  density is high

21
Related work

• Measurement methodology
• IBM (Shaikh et al.)
• Time correlation of DNS and HTTP requests from
  DNS and Web server logs
• Univ of Boston (Bestavros et al.)
• Assigning multiple IP addresses to a Web server
• Differences from our work
• Our methodology efficient, accurate,
  nonintrusive
• Web bugs
• Proximity metrics
• Ciscos Boomerang protocol uses latency from
  cache servers to the LDNS