International Business Machines Corporation

Josyula R Rao

Suresh N Chari

Dimitrios Pendarakis

Pankaj Rohatgi

Ian M Molloy

Marc Philippe Stoecklin

Douglas Lee Schales

Reiner Sailer

Chung-Sheng Li

Pratap Chandra Pattnaik

Youngja Park

Hubertus Franke

Nagui Halim

Matt Richard Hogstrom

Brad L Brech

Michael D Williams

Wilfried Teiken

Scott W Crowder

Helmut Scherzer

Radha P Ratnaparkhi

Ashish Kundu

Xin Hu

Pau-Chen Cheng

Mihai Christodorescu

Ted A Habeck

Jiyong Jang

Vincenzo Valentino DiLuoffo

Taesung Lee

Elaine Rivette Palmer

Paul Ashley Karger

William Alexander Bird

Stefan Berger

Jialong Zhang

Dhilung Hang Kirat

Zhongshu Gu

Michael Steiner

Heqing Huang

Tal D Rabin

Suzanne Carol Deffeyes

Bruce Roy Archambeault

David Claude Toll

Claudia Keser

Yangyi Chen

Wietse Z Venema

Rick A Hamilton, Ii

Dakshi Agrawal

Vijay A Kumar

Lawrence H Koved

Charanjit Singh Jutla

Youssef Drissi

Srinivasan Parthasarathy

Lisa Amini

Kapil Kumar Singh

Colin George Harrison

Pradeep Kumar Dubey

Donald Philip Pazel

Mitchell Adam Cohen

Stephen Carl Gates

Jack Lawrence Kouloheris

Michael S Meier

Kevan Lee Miller

Oliver Verscheure

Atri Rudra

Matthias Kaiserwerth

Dimitrious Pendarakis

James Robert Russell

Willliam J Rippon

Mechanisms are provided for detecting abnormal system call sequences in a monitored computing environment. The mechanisms receive, from a computing system resource of the monitored computing environment, a system call of an observed system call sequence for evaluation. A trained recurrent neural network (RNN), trained to predict system call sequences, processes the system call to generate a prediction of a subsequent system call in a predicted system call sequence. Abnormal call sequence logic compares the subsequent system call in the predicted system call sequence to an observed system call in the observed system call sequence and identifies a difference between the predicted system call sequence and the observed system call sequence based on results of the comparing. The abnormal call sequence logic generates an alert notification in response to identifying the difference.

Recurrent neural network based anomaly detection

An automated method for processing security events in association with a cybersecurity knowledge graph. The method begins upon receipt of information from a security system representing an offense. An initial offense context graph is built based in part on context data about the offense. The graph also activity nodes connected to a root node; at least one activity node includes an observable. The root node and its one or more activity nodes represent a context for the offense. The knowledge graph, and potentially other data sources, are then explored to further refine the initial graph to generate a refined graph that is then provided to an analyst for further review and analysis. Knowledge graph exploration involves locating the observables and their connections in the knowledge graph, determining that they are associated with known malicious entities, and then building subgraphs that are then merged into the initial graph.

Cognitive offense analysis using contextual data and knowledge graphs

Detecting malicious user activity is provided. A profile for a user that accesses a set of protected assets is generated based on static information representing an organizational view and associated attributes corresponding to the user and based on dynamic information representing observable actions made by the user. A plurality of analytics is applied on the profile corresponding to the user to generate an aggregate risk score for the user accessing the set of protected assets based on applying the plurality of analytics on the profile of the user. A malicious user activity alert is generated in response to the aggregate risk score for the user accessing the set of protected assets being greater than an alert threshold value. The malicious user activity alert is sent to an analyst for feedback.

Detecting malicious user activity

There is a computer program product and computer system that includes program instructions programmed to establish a security container describing a workload and a set of resources in a software-defined environment, the security container including a set of sub-containers that are self-describing sub-containers having associated metadata describing content of a respectively corresponding sub-container, each sub-container of the set of sub-containers respectively corresponds to a resource-divisible portion of the workload, the set of resources being required by the workload, wherein a sub-container of the set of sub-containers is an operating system sub-container; monitor the workload and the set of resources for security events; and responsive to identifying a security event, adjust isolation mechanisms provided by the plurality of sub-containers at various layers of a stack. The set of sub-containers represents an end-to-end run time environment for processing the workload using the set of resources.

Security within a software-defined infrastructure

There is a computer program product and computer system that includes program instructions programmed to identify, in a software-defined environment, a security container describing a workload and a set of resources required by the workload, the security container including self-describing sub-containers having associated metadata describing content of a respectively corresponding sub-container; determine, for the workload, a set of resource-divisible portions of the workload including a compute-resource portion; generate a plurality of sub-containers within the security container, a sub-container within the plurality of sub-containers being a self-describing sub-container having associated metadata describing the content of the sub-container representing only one resource-divisible portion, the sub-container being an operating system sub-container; and responsive to identifying a security event while processing the workload, adjust a security mechanism associated with the security container. The plurality of sub-containers represents an end-to-end run time environment for processing the workload.

Generating an attack graph to protect sensitive data objects from attack is provided. The attack graph that includes nodes representing components in a set of components of a regulated service and edges between nodes representing relationships between related components in the set of components is generated based on vulnerability and risk metrics corresponding to each component. A risk score is calculated for each component represented by a node in the attack graph based on sensitivity rank and criticality rank corresponding to each respective component. Risk scores are aggregated for each component along each edge path connecting a node of a particular component to a node of a related component. In response to determining that an aggregated risk score of a component is greater than or equal to a risk threshold, an action is performed to mitigate a risk to sensitive data corresponding to the component posed by an attack.

Automatic generation of data-centric attack graphs

Log(s) of IT events are accessed in a distributed system that includes a distributed application. The distributed system includes multiple data objects. The distributed application uses, processes, or otherwise accesses one or more of data objects. The IT events concern the distributed application and concern accesses by the distributed application to the data object(s). The IT events are correlated with a selected set of the data objects. Risks are estimated to the selected set of data objects based on the information technology events. Estimating risks uses at least ranks of compliance rules as these rules apply to the data objects in the system, and vulnerability scores of systems corresponding to the set of data objects and information technology events. Information is output that allows a user to determine the estimated risks for the selected set of data objects. Techniques for determining ranks of compliance rules are also disclosed.

Data-centric monitoring of compliance of distributed applications

A method includes collecting system calls and call parameters invoked by monitored applications for target computer systems. The system calls and call parameters are received from operating system kernels on the plurality of target computer systems. Sequences of systems calls and call parameters of the monitored applications are correlated among different target computer systems to deduce malicious activities. Remedial action(s) are performed in response to malicious activities being deduced as being malicious by the correlating. Another method includes determining that network activity at a specific time is deemed to be suspicious. Using IP addresses involved in the suspicious network activity, computer system(s) are determined that are sources of the suspicious network activity. Based on the specific time and the determined computer system(s), application(s) are determined that are executing on the determined computer system(s) that are causing the suspicious network activity. Remedial action(s) are performed for the determined computer system(s).

Computer implemented techniques for detecting, investigating and remediating security violations to IT infrastructure

Generating an attack graph is provided. A set of sensitive data corresponding to a regulated service is identified. A set of components corresponding to the regulated service that are authorized to perform activities associated with sensitive data is scanned for. Vulnerability and risk metrics corresponding to each component in the set of components of the regulated service is identified. The attack graph that includes nodes representing components in the set of components of the regulated service and edges between nodes representing relationships between related components in the set of components is generated based on the vulnerability and risk metrics corresponding to each component in the set of components.

There is a method and system that includes establishing a security container that describes a workload and a set of resources that corresponds to the workload in a software-defined environment, determining a set of security criteria for the security container, monitoring the workload and the set of resources for security events based, at least in part, upon the set of security criteria, and responsive to identifying a security event, adjusting one or more security mechanisms. The steps of monitoring and adjusting are operated within the software-defined environment.

Embodiments include a network data collection and response system for enhancing security in an enterprise network providing a user-supplied computing device with access to the network. A network data collection and response system tracks network activity of the device and maintains a device inventory recording the device type and configuration information for the device along with a resource utilization profile for the device. The network data collection and response system detects high-risk or unauthorized network activity involving the device through passive monitoring without utilization of a data monitoring agent installed on the device and implements a response action to mitigate the high-risk or unauthorized network.

Network data collection and response system

Growing community of inventors

Briarcliff Manor, NY, United States of America

Josyula R Rao