Abstract: An example method can include choosing a pattern or patterns of network traffic. This pattern can be representative of a certain type of traffic such as an attack. The pattern can be associated with various components of a network and can describe expected behavior of these various components. A system performing this method can then choose a nodes or nodes to generate traffic according to the pattern and send an instruction accordingly. After this synthetic traffic is generated, the system can compare the behavior of the components with the expected behavior. An alert can then be created to notify an administrator or otherwise remedy any problems.

Abstract: An example method includes detecting, using sensors, packets throughout a datacenter. The sensors can then send packet logs to various collectors which can then identify and summarize data flows in the datacenter. The collectors can then send flow logs to an analytics module which can identify the status of the datacenter and detect an attack.

Abstract: Systems, methods, and computer-readable media are provided for determining whether a node in a network is a server or a client. In some examples, a system can collect, from one or more sensors that monitor at least part of data traffic being transmitted via a pair of nodes in a network, information of the data traffic. The system can analyze attributes of the data traffic such as timing, port magnitude, degree of communication, historical data, etc. Based on analysis results and a predetermined rule associated with the attributes, the system can determine which node of the pair of nodes is a client and which node is a server.

Abstract: This disclosure generally relate to a method and system for mapping network information. The present technology relates techniques that enable full-scale, dynamic network mapping of a network system. By collecting network and computing data using built-in sensors, the present technology can provide network information for system monitoring and maintenance. According to some embodiments, the present technology enables generating and displaying of network connections and data processing statistics related to numerous nodes in a network. The present technology provides useful insights and actionable knowledge for network monitoring, security, and maintenance, via intelligently summarizing and effectively displaying the complex network communications and processes of a network.

Abstract: Systems, methods, and computer-readable media are provided for determining whether a node in a network is a server or a client. In some examples, a system can collect, from one or more sensors that monitor at least part of data traffic being transmitted via a pair of nodes in a network, information of the data traffic. The system can analyze attributes of the data traffic such as timing, port magnitude, degree of communication, historical data, etc. Based on analysis results and a predetermined rule associated with the attributes, the system can determine which node of the pair of nodes is a client and which node is a server.

Abstract: Systems, methods, and computer-readable media are provided for determining a packet's round trip time (RTT) in a network. A system can receive information of a packet sent by a component of the network and further determine an expected acknowledgement (ACK) sequence number associated with the packet based upon received information of the packet. The system can receive information of a subsequent packet received by the component and determine an ACK sequence number and a receiving time of the subsequent packet. In response to determining that the ACK sequence number of the subsequent TCP packet matches the expected ACK sequence number, the system can determine a round trip time (RTT) of the packet based upon the received information of the packet and the received information of the subsequent packet.

Abstract: Systems and methods are provided for automatically discovering applications/clusters in a network and mapping dependencies between the applications/clusters. A network monitoring system can capture network flow data using sensors executing on physical and/or virtual servers of the network and sensors executing on networking devices connected to the servers. The system can determine a graph including nodes, representing at least the servers, and edges, between pairs of the nodes of the graph indicating the network flow data includes one or more observed flows between pairs of the servers represented by the pairs of the nodes. The system can determine a dependency map, including representations of clusters of the servers and representations of dependencies between the clusters, based on the graph. The system can display a first representation of a first cluster of the dependency map and information indicating a confidence level of identifying the first cluster.

Abstract: Application dependency mapping (ADM) can be automated in a network. The network can determine an optimum number of clusters for the network using the minimum description length principle (MDL). The network can capture network and associated data using a sensor network that provides multiple perspectives and generate a graph therefrom. The nodes of the graph can include sources, destinations, and destination ports identified in the captured data, and the edges of the graph can include observed flows from the sources to the destinations at the destination ports. Each clustering can be evaluated according to an MDL score. The optimum number of clusters for the network may correspond to the number of clusters of the clustering associated with the minimum MDL score.

Abstract: An example method includes detecting, using sensors, packets throughout a datacenter. The sensors can then send packet logs to various collectors which can then identify and summarize data flows in the datacenter. The collectors can then send flow logs to an analytics module which can identify the status of the datacenter and detect an attack.

Abstract: Application dependency mapping can be automated in a network. The network can capture traffic data for flows passing through the network using a sensor network that provides multiple perspectives for the traffic. The network can analyze the traffic data to identify endpoints of the network. The network can also identify particular network configurations from the traffic data, such as a load balancing schema or a subnetting schema. The network can partition the endpoints based on the network configuration(s) and perform similarity measurements of endpoints in each partition to determine clusters of each partition. The clusters can make up nodes of an application dependency map, and relationships between and among the clusters can make up edges of the application dependency map.

Abstract: Flow data can be augmented with features or attributes from other domains, such as attributes from a source host and/or destination host of a flow, a process initiating the flow, and/or a process owner or user. A network can be configured to capture network or packet header attributes of a first flow and determine additional attributes of the first flow using a sensor network. The sensor network can include sensors for networking devices (e.g., routers, switches, network appliances), physical servers, hypervisors or container engines, and virtual partitions (e.g., virtual machines or containers). The network can calculate a feature vector including the packet header attributes and additional attributes to represent the first flow. The network can compare the feature vector of the first flow to respective feature vectors of other flows to determine an applicable policy, and enforce that policy for subsequent flows.

Abstract: An example method includes detecting, using sensors, packets throughout a datacenter. The sensors can then send packet logs to various collectors which can then identify and summarize data flows in the datacenter. The collectors can then send flow logs to an analytics module which can identify the status of the datacenter and detect an attack.

Abstract: Systems, methods, and computer-readable media are provided for determining a packet's round trip time (RTT) in a network. A system can receive information of a packet sent by a component of the network and further determine an expected acknowledgement (ACK) sequence number associated with the packet based upon received information of the packet. The system can receive information of a subsequent packet received by the component and determine an ACK sequence number and a receiving time of the subsequent packet. In response to determining that the ACK sequence number of the subsequent TCP packet matches the expected ACK sequence number, the system can determine a round trip time (RTT) of the packet based upon the received information of the packet and the received information of the subsequent packet.

Abstract: This disclosure generally relates to a method and system for generating a communication graph of a network using an application dependency mapping (ADM) pipeline. In one aspect of the disclosure, the method comprises receiving network data (e.g., flow data and process information at each node) from a plurality of sensors associated with a plurality of nodes of the network, determining a plurality of vectors and an initial graph of the plurality of nodes based upon the network data, determining similarities between the plurality of vectors, clustering the plurality of vectors into a plurality of clustered vectors based upon the similarities between the plurality of vectors, and generating a communication graph of the network system based upon the plurality of clustered vectors.

Abstract: This disclosure generally relate to a method and system for mapping network information. The present technology relates techniques that enable full-scale, dynamic network mapping of a network system. By collecting network and computing data using built-in sensors, the present technology can provide network information for system monitoring and maintenance. According to some embodiments, the present technology enables generating and displaying of network connections and data processing statistics related to numerous nodes in a network. The present technology provides useful insights and actionable knowledge for network monitoring, security, and maintenance, via intelligently summarizing and effectively displaying the complex network communications and processes of a network.

Abstract: Application dependency mapping (ADM) can be automated in a network. The network can determine whether certain nodes form a cluster of a tier of an application. The network can monitor network data and process data for traffic passing through the network using a sensor network that provides multiple perspectives for the traffic. The network can analyze the network data and process data to determine respective feature vectors for nodes. A feature vector may represent a combination of the features corresponding to the network data and the features corresponding to the process data of a node. The network can compare the similarity of the respective feature vectors and determine each node's cluster based on similarity measures between nodes.

Abstract: Application dependency mapping can be automated in a network. The network can capture traffic data for flows passing through the network using a sensor network that provides multiple perspectives for the traffic. The network can analyze the traffic data to identify endpoints of the network. The network can also identify particular network configurations from the traffic data, such as a load balancing schema or a subnetting schema. The network can partition the endpoints based on the network configuration(s) and perform similarity measurements of endpoints in each partition to determine clusters of each partition. The clusters can make up nodes of an application dependency map, and relationships between and among the clusters can make up edges of the application dependency map.

Abstract: Flow data can be augmented with features or attributes from other domains, such as attributes from a source host and/or destination host of a flow, a process initiating the flow, and/or a process owner or user. A network can be configured to capture network or packet header attributes of a first flow and determine additional attributes of the first flow using a sensor network. The sensor network can include sensors for networking devices (e.g., routers, switches, network appliances), physical servers, hypervisors or container engines, and virtual partitions (e.g., virtual machines or containers). The network can calculate a feature vector including the packet header attributes and additional attributes to represent the first flow. The network can compare the feature vector of the first flow to respective feature vectors of other flows to determine an applicable policy, and enforce that policy for subsequent flows.

Abstract: This disclosure generally relate to a method and system for mapping application dependency information. The present technology relates techniques that enable user-adjustable application dependency mapping of a network system. By collecting internal network data using various sensors in conjunction with external user inputs, the present technology can provide optimized application dependency mapping using user inputs.

Abstract: An example method includes detecting, using sensors, packets throughout a datacenter. The sensors can then send packet logs to various collectors which can then identify and summarize data flows in the datacenter. The collectors can then send flow logs to an analytics module which can identify the status of the datacenter and detect an attack.