Relative Network Positioning via CDN Redirections AoJan Su

Relative Network Positioning via CDN Redirections Ao-Jan Su, David R. Choffnes, Fabián E. Bustamante and Aleksandar Kuzmanovic Department of EECS Northwestern University IEEE ICDCS 2008

Network Positioning Why do we need network positioning systems? – Emerging large scale distributed systems can benefit from selecting among alternative nodes • Example: select an on-line gaming server – All-to-all measurements are not scalable Current approaches – Provide network positioning services in a scalable way (e. g. landmark based) – Clear tradeoffs • Precision vs. overhead • Precision vs. deployment • And others Ao-Jan Su Relative Network Positioning via CDN Redirections 2

Observations Content distribution networks (e. g. , Akamai) improve web performance by – Performing extensive network measurements – Redirecting clients to their closest replica servers – Publishing the results through DNS Can we reuse those measurements collected by CDNs to build a network positioning system? Ao-Jan Su Relative Network Positioning via CDN Redirections 3

CDNs Basics Web client’s request redirected to ‘close’ by server – Client gets web site’s DNS CNAME entry with domain name in CDN network – Hierarchy of CDN’s DNS servers direct client to 2 nearby servers Hierarchy of CDN DNS servers Internet Customer DNS servers Multiple redirections to find nearby edge servers Web replica servers (3) (4) Client is given 2 nearby web (2) Client gets CNAME entryservers (fault replica tolerance) with domain name in Akamai Client requests translation for yahoo Ao-Jan Su LDNS (5) (6) (1) Relative Network Positioning via CDN Redirections Web client 4

Our approach CDN-based Relative Network Positioning (CRP) Clients are redirected to currently closest replica servers in general CDN’s redirections are primarily driven by network conditions (latency) [Su et al. 2006] Inferring relative network distance by overlapping CDN replica servers R 1 A Ao-Jan Su R 2 B Relative Network Positioning via CDN Redirections C 5

Uses of CRP Closest node selection – Select the closest node (shortest latency) from a group of candidates (e. g. select the closest on-line gaming server) – Methodology • Encode redirection frequency from a node to its redirected replica servers by a vector • Compare similarity (cosine similarity) of nodes’ redirection vectors to estimate proximity Replica servers R 1 Server A 0. 8 0. 2 Client Ao-Jan Su R 2 Server B R 3 Relative Network Positioning via CDN Redirections 6

Uses of CRP (Cont. ) Clustering – Select a set of nodes that are close to each other (e. g. replicate content to a group of nodes) – Methodology • Select cluster centers • Assign strong mapping peers to the cluster centers Ao-Jan Su Relative Network Positioning via CDN Redirections 7

Evaluation Goals Comparing the performance of CRP’s closest node selection to – Ground truth – active measurements – A state of the art network positioning system – Meridian [Wong 2005] With respect to – – Accuracy Scalability Deployment Overhead Meridian: Closest node selection Ao-Jan Su Relative Network Positioning via CDN Redirections 8

Experiment Setup Globally distributed nodes – 1000 DNS servers as clients – 240 Planet Lab nodes as candidate servers (on the same nodes as our reference system – Meridian) Concurrent data collection – Monitoring CDN redirections by recursive DNS queries for CRP – Querying Meridian via its interface – Measuring end-to-end latencies by pings as the ground truth Ao-Jan Su Relative Network Positioning via CDN Redirections 9

Selecting the closest node Clients are not close to any servers due to • Limited Planet Lab nodes coverage (Meridian) • Located in areas not well served by CDNs (CRP) CRP outperforms Meridian by 25% of the nodes due to larger deployment of CDN replica servers to its alternative CRP’s accuracy is comparable CRP’s recommendations for 65% of nodes differ frommeasurements Meridian by < 7 ms without active and dedicated infrastructure Ao-Jan Su Relative Network Positioning via CDN Redirections 10

Selecting the closest node (Cont. ) Relative Error: estimated latency – ground truth 80% of CRP nodes comparing have CRP’s is quite accurate to ground truth, relative error < 50 ms with virtually no measurement overhead Ao-Jan Su Relative Network Positioning via CDN Redirections 11

Load on CDN’s DNS System Rank: rank 0 is the closest server Low probe frequency • Smaller overhead • Less accurate • Miss overlapping replica servers 100 mins probe frequency High probe frequency • Appropriate for 95%Improve of nodes • Can accuracy • Much less than CDN’s DNS TTL (20 secs) • Larger overhead • Overhead is too small to impact CDN’s operations Ao-Jan Su Relative Network Positioning via CDN Redirections 12

Load on CRP Clients Small history Large history • More refined results • Larger computation overhead Ao-Jan Su • Sufficient for CRP • Small overhead • Capture network dynamics Relative Network Positioning via CDN Redirections 13

You May Be Wondering “Will CDNs be unhappy because of CRP? ” – CRP nodes behaves as regular web clients – CRP’s overhead does not impact CDN’s daily operations – Could be an additional service provided by CDNs “What if CDNs change their redirection policy? ” – CRP’s goal aligns with CDNs – Our approach is not restricted to a specific CDN, CRP can reuse results from other measurement infrastructures Ao-Jan Su Relative Network Positioning via CDN Redirections 14

Summary CRP discovers relative positions of end hosts – Closest node selection – Clustering Key features of CRP – Accurate – Light-weight • Reuse CDN’s network measurements – Scalable • No dedicated infrastructure is required Ao-Jan Su Relative Network Positioning via CDN Redirections 15

Cosine Similarity Ao-Jan Su Relative Network Positioning via CDN Redirections 16