公开(公告)号：US11528288B2

公开(公告)日：2022-12-13

申请号：US16691475

申请日：2019-11-21

申请人： OVH

发明人： Olivier Nicol

IPC分类号： H04L9/40 ,  G06F16/903

摘要： Methods allow a predicting and detecting potential anomalies at a service infrastructure. A strings table having entries that define character strings and corresponding anomaly probabilities is accessed. A log entry related to an event occurring in the service infrastructure is generated in a database. The log entry includes a character string designating a name of a file or an IP address and a domain name hosted by the service infrastructure. A search is made for the character string in the strings table. The domain name is marked as suspect if the character string is found in the strings table and if an anomaly probability for the character string exceeds a predetermined threshold. The anomaly probabilities may be calculated using a Bayesian filter that accounts for a number of domains hosted by the service infrastructure on which the character string has recently appeared.

公开(公告)号：US11288260B2

公开(公告)日：2022-03-29

申请号：US17306265

申请日：2021-05-03

申请人： OVH

发明人： Olivier Nicol ,  Laurent Parmentier ,  Pauline Wauquier

IPC分类号： G06F16/23 ,  G06F17/18

摘要： A method and system for detecting anomalies in a data pipeline are disclosed. Data tables are extracted from data pipelines. A best time column is calculated for each data table. Table lines are aggregated so as to establish counts of data lines per a time interval, and corresponding sequences of points are produced that are being analysed through regression methods. An anomaly is raised when points have counts of data lines that lie outside confidence intervals around expected values for such counts.