A device receives network data associated with a network that includes network devices interconnected by links at an Internet protocol (IP) layer and an optical layer of the network. The device receives constraints associated with determining a network plan for the network, where the constraints include a constraint indicating a particular time period associated with determining potential network plans for the network. The device identifies variables and values of the variables for the network plan based on the network data, and determines, within the particular time period, the potential network plans for the network based on the constraints and the values of the variables. The device identifies a potential network plan, of the potential network plans, that minimizes costs associated with operating the network, and causes the identified potential network plan to be implemented in the network by the network devices.

Techniques are described herein for generating network topologies based on models, and deploying the network topologies across hybrid clouds and other computing environments that include multiple workload resource domains. A topology deployment system may receive data representing a logical topology model, and may generate a network topology for deployment based on the logical model. The network topology may include various services and/or other resources provided by different tenants in the computing environment, and tenant may be associated with different set of resources and deployment constraints. The topology deployment system may determine and generate the network topology to use the various resources and comply with various deployment constraints of the different tenants providing the services, and the tenants consuming the network topology.

A computer-implemented method upgrades an edge network based on analysis by a learning model. The method includes identifying, in a network, a plurality of devices, where each device in the network is configured to provide data on at least one other device in the network. The method also includes determining capabilities of each device of the plurality of devices. The method further includes monitoring, for each device, capacity information and tasks performed during operation of the network. The method includes analyzing, based on the monitoring, each use of each device. The method also includes recommending, in response to the analyzing and by a learning model, a first upgrade to the network. The method further includes implementing the first upgrade.

Methods and systems for partially or fully distributed network verification are described. In partially distributed network verification, each network device generates a respective device-level binary decision diagram (BDD) representing the logical behavior of the respective network device for a network property of interest. The device-level BDDs from each network device are received by a verification service that performs verification by generating an input BDD representing an input header space, and applies each device-level BDD in a logical path from a source device to a destination device, and reports the output BDD. In fully distributed network verification, each network device is responsible for calculating a device-specific output BDD by applying a device-specific BDD, which represents the logical behavior of the network device, to a device-specific input BDD.

A cloud network is a complex environment in which hundreds and thousands of users or entities can each host, create, modify, and develop multiple virtual machines. Each virtual machine can have complex behavior unknown to the provider or maintainer of the cloud. Technologies disclosed include methods, systems, and apparatuses to monitor the complex environment to detect network anomalies using machine learning techniques. In addition, techniques to modify and adapt to user feedback are provided allowing the developed models to be tuned for specific use cases, virtual machine types, and users.

Disclosed are systems, methods, and computer-readable media for assuring tenant forwarding in a network environment. Network assurance can be determined in layer 1, layer 2 and layer 3 of the networked environment including, internal-internal (e.g., inter-fabric) forwarding and internal-external (e.g., outside the fabric) forwarding in the networked environment. The network assurance can be performed using logical configurations, software configurations and/or hardware configurations.

Disclosed is a system, method, and computer program product for implementing a log analytics method and system that can configure, collect, and analyze log records in an efficient manner. An improved approach is provided for identifying log files that have undergone a change in status that would require retrieve of its log data, by including a module directly into the operating system that allows the log collection component to be reactively notified of any changes to pertinent log files.

A procedure for evaluating a network, and a system, apparatus, and computer program that operate in accordance with the procedure. The procedure includes aggregating packet information from one or more sources in a network, and executing a correlation algorithm to determine traffic flow information based on the packet information. The aggregating includes obtaining information from a header of a packet being communicated in the network, in one example embodiment. In another example, the executing includes tracing a traffic flow from a source node to a destination node, and the tracing includes determining, based on the packet information, each link by which the traffic flow is communicated from the source node to the destination node.

The present disclosure relates to an apparatus and method of altruistic scheduling based on reinforcement learning. An altruistic scheduling apparatus according to an embodiment of the present disclosure includes: an external scheduling agent for determining a basic resource share for each process based on information of a resource management system; an internal scheduling agent for determining a basic resource allocation schedule for each process based on information including the basic resource share and a resource leftover based on the basic resource allocation schedule; and a leftover scheduling agent for determining a leftover resource allocation schedule based on information including the resource leftover. According to an embodiment of the present disclosure, it may be expected that reinforcement learning will not only mitigate the diminution of fairness of an altruistic scheduler but also further improve other performance indicators such as completion time and efficiency.

A kind of transmission method based on the learnable power model, which conducts periodic record for the historical change trend of the network. This method conducts weighting smooth processing on the round trip time and judges the changing trend of congestion control window. Then, it establishes model for the relationship between network power and the congestion control widow. When a new ACK is received, it immediately updates the window of power model. Finally, it forecasts the size of the congestion control window of the next time period by combining the congestion window and the network power changing trend. For the network packet loss or time-out events, the retransmission mechanism of traditional TCP is used, and when the packet loss ends, the power model process is used again. This invention reduces the influence of the network random events of the estimation error of traditional algorithm.