Abstract:
A method is provided that includes transforming training data into a neural network based learning model using a set of temporal graphs derived from the training data. The method includes performing model learning on the learning model by automatically adjusting learning model parameters based on the set of the temporal graphs to minimize differences between a predetermined ground-truth ranking list and a learning model output ranking list. The method includes transforming testing data into a neural network based inference model using another set of temporal graphs derived from the testing data. The method includes performing model inference by applying the inference and learning models to test data to extract context features for alerts in the test data and calculate a ranking list for the alerts based on the extracted context features. Top-ranked alerts are identified as critical alerts. Each alert represents an anomaly in the test data.
Abstract:
Methods and systems for detecting security intrusions include detecting alerts in monitored system data. Temporal dependencies are determined (306) between the alerts based on a prefix tree formed from the detected alerts. Content dependencies between the alerts are determined (308) based on a distance between alerts in a graph representation of the detected alerts. The alerts are ranked (310) based on an optimization problem that includes the temporal dependencies and the content dependencies. A security management action (614) is performed based on the ranked alerts.
Abstract:
A heterogeneous log pattern editing recommendation system and computer- implemented method are provided. The system (600) has a processor (605) configured to identify, from heterogeneous logs, patterns including variable fields and constant fields. The processor (605) is also configured to extract a category feature, a cardinality feature, and a before-after n-gram feature by tokenizing the variable fields in the identified patterns. The processor (605) is additionally configured to generate target similarity scores between target fields to be potentially edited and other fields from among the variable fields in the heterogeneous logs using pattern editing operations based on the extracted category feature, the extracted cardinality feature, and the extracted before-after n-gram feature. The processor (605) is further configured to recommend, to a user, log pattern edits for at least one of the target fields based on the target similarity scores between the target fields in the heterogeneous logs.
Abstract:
Systems and methods are disclosed for parsing logs from arbitrary or unknown systems or applications by capturing heterogeneous logs from the arbitrary or unknown systems or applications; generating one pattern for every unique log message; building a pattern hierarchy tree by grouping patterns based on similarity metrics, and for every group it generates one pattern by combing all constituting patterns of that group; and selecting a set of patterns from the pattern hierarchy tree.
Abstract:
Methods and systems for detecting malicious processes include modeling system data as a graph comprising vertices that represent system entities and edges that represent events between respective system entities. Each edge has one or more timestamps corresponding respective events between two system entities. A set of valid path patterns that relate to potential attacks is generated. One or more event sequences in the system are determined to be suspicious based on the graph and the valid path patterns using a random walk on the graph.
Abstract:
A method and system for constructing behavior queries in temporal graphs using discriminative sub-trace mining. The method (100) includes generating system data logs to provide temporal graphs (102), wherein the temporal graphs include a first temporal graph corresponding to a target behavior and a second temporal graph corresponding to a set of background behaviors (102), generating temporal graph patterns for each of the first and second temporal graphs to determine whether a pattern exists between a first temporal graph pattern and a second temporal graph pattern, wherein the pattern between the temporal graph patterns is a non-repetitive graph pattern (104), pruning the pattern between the first and second temporal graph patterns to provide a discriminative temporal graph (106), and generating behavior queries based on the discriminative temporal graph (110).
Abstract:
Systems and methods for controlling legacy switch routing in one or more hybrid networks of interconnected computers and switches, including generating a network underlay (304) for the one or more hybrid networks by generating a minimum spanning tree (MST) (306) and a forwarding graph (FWG) (308) over a physical network topology of the one or more hybrid networks (400), determining an optimal path between hosts on the FWG by optimizing an initial path with a minimum cost mapping (312), and adjusting the initial path (310) to enforce the optimal path (314) by generating and installing special packets in one or more programmable switches to trigger installation of forwarding rules for one or more legacy switches (516).
Abstract:
Systems and methods for decoupled searching and optimization for one or more data centers, including determining a network topology for one or more networks of interconnected computer systems embedded in the one or more data centers (304), searching for routing candidates based on a network topology determined (310), and updating (314) and applying (316) one or more objective functions to the routing candidates to determine an optimal routing candidate to satisfy embedding goals based on tenant requests, and to embed the optimal routing candidate in the one or more data centers (412).
Abstract:
Systems and methods are disclosed to schedule virtual machine (VM) migrations by analyzing VM migration behavior; building a simulation tool to predict time for multiple migrations under different links conditions and VM characteristics; determing a predetermined bandwidth sharing policy for each network link; applying a bin-packing technique to organize bandwidth resources from all network links, and allocating the links to different migration tasks.
Abstract:
A method and system for predicting the performance of a multi-tier computer software system operating on a distributed computer system, sends client requests to one or more tiers of software components of the multi-tier computer software system in a time selective manner; collects traffic traces among all the one or more tiers of the software components of the multi-tier computer software system; collects CPU time at the software components of the multi-tier computer software system; infers performance data of the multi-tier computer software system from the collected traffic traces; and determines disk input/output waiting time from the inferred performance data.