Pegasus: Precision Hunting for Icebergs and Anomalies work Flows Sriharsha Gangam ?, Puneet Sharma +, Sonia Fahmy ??Purdue University, +HP Labs E-mail: ******@, @, ******@ Abstract—Accurate work monitoring is crucial for detecting attacks, faults, and anomalies, and determiningtraf?c properties across work. With high bandwidth links and consequently increasing traf?c volumes, it is dif?cult to collect and analyze detailed ?ow records in an online manner. Tra- ditional solutions that decouple data collection from analysis resort to sampling and sketching to handle large monitoring traf?c volumes. We propose a new system, Pegasus, to mercially available co-pute and storage devices near routers and switches. Pegasusadaptivelymanages data transfers between monitors and aggregators based on traf?c patterns and user queries. We use Pegasus to detectglobal lobal heavy-hitters. Icebergs are ?ows with mon property that contribute a signi?cant fraction work traf?c. For example, DDoS attack detection is an iceberg detection problem with mon destination IP. Other applications include identi?cationof “top talkers,” top destinations, and detection of worms and portscans. Experiments with Abilene traces, sFlow traces froman work, and deployment of Pegasus as a live monitoring service on Lab show that our system is accurate and scales well with increasing tr
Pegasus Precision Hunting for Icebergs and 来自淘豆网m.daumloan.com转载请标明出处.
