International Business Machines Corporation

Michael John Suzio

Gary I Givental

HuyAnh Dinh Ngo

Daniel E Chapman, Ii

John D Kuhn

An event retrieval and analysis system compares counts of event data for a device to stored profile counts to determine if alerts should be triggered. Event data can be retrieved by a sensor. Rules for analyzing the event data can be retrieved based on the device. The event data is analyzed based on the rules to determine recordable events. Recordable events are organized into categories representing a type or severity of attack. Current event counts are calculated by summing the recordable events for each category. A normal profile is retrieved for the device and compared to the current event count. A percentage change trigger can be retrieved from a threshold matrix based on the current event count. The percentage increase of the current event count over the normal profile is calculated and compared to the percentage change trigger to determine if an alert is triggered by the analysis system.

Method and system for analysis of security events in a managed computer network

A computer detects malicious intrusions (or bots) into a computer. The computer receives firewall log data that includes communication records containing the source and destination of the communication, as well as, the time of the communication. The source or destination of the communication may be on a list of suspicious servers known to contain malicious software. The computer identifies a sequence of communications between a common source address and a common destination address. The computer further identifies substantially fixed intervals between the communications, and generates an alert indicating a suspected bot intrusion. The computer also identifies from the sequence of communication, patterns in the communication intervals, similarly generating an alert indicating a suspected bot intrusion.

Growing community of inventors

Royal Oak, MI, United States of America

Michael John Suzio