An example network orchestrator includes processing circuitry and a memory. The memory includes instructions that cause the network orchestrator to receive network probe information including delay times of network probes associated with a set of flows between devices. The instructions further cause the network orchestrator to generate a correlation matrix including correlations representing shared congested links between pairs of flows. The instructions further cause the network orchestrator to for each flow of the set of flows, determine a routing solution optimized for the each flow and select a total minimum cost solution from the determined routing solutions.

Techniques for dynamic resource movement in heterogeneous computing environments including provider substrate extensions are described. A dynamic resource movement service of a provider network monitor conditions of heterogeneous computing environments, including provider substrate extensions of the cloud provider network, to evaluate customer-provided movement policy conditions governing when to move customer application resources from these environments, where to move the resource to, and/or how to move the customer application resources. The customer-provided movement policy conditions may be based on a variety of factors, such as a latency between end-users of the customer application and the application itself.

A method and apparatus form and/or define a network topology in a Layer 3 network with a plurality of nodes, where each node has at least one interface. To that end, the method defines a plurality of neighborhoods, and assigns at least one interface of each node to at least one of the neighborhoods. The method also assigns a communication role to each interface so that each communication role is effective relative to one of the plurality of neighborhoods. The method then enables communication between the interfaces of the plurality of nodes as a function of the neighborhoods and the communication roles.

Some embodiments provide a method of selecting data links for an application in a network. The method receives, from a machine implementing the application, a set of identifiers of required link characteristics. Based on at least one of the identifiers, the method selects a transport group that includes a set of optional links matching the identifiers. From the selected transport group, the method selects a link matching the set of identifiers.

Some embodiments provide a method of selecting data links for an application in a network. The method receives, from a machine implementing the application, a set of identifiers of required link characteristics. Based on at least one of the identifiers, the method selects a transport group that includes a set of optional links matching the identifiers. From the selected transport group, the method selects a link matching the set of identifiers.

Various techniques for dynamic path selection and data flow forwarding are disclosed. For example, various systems, processes, and computer program products for dynamic path selection and data flow forwarding are disclosed for providing dynamic path selection and data flow forwarding that can facilitate preserving/enforcing symmetry in data flows as disclosed with respect to various embodiments.

The present disclosure provides a logical node distributed signature decision system for a distributed data processing system, including: an initial logical node generating assembly, configured to receive task configuration data input by a user, and generate an initial logical node topology for the distributed data processing system, wherein a source logical node has a specified logical distributed signature, each initial logical node is attached with a candidate logical distributed signature set based on the task configuration data; and a logical distributed signature selecting assembly, configured to, according to a distributed descriptor of an output end of each upstream logical node for which a logical distributed signature is already determined, for each candidate logical distributed signature of a current logical node, compute a cost of data transmission required to transform the distributed descriptor of the tensor of the output end of each upstream logical node into the distributed descriptor.

Some embodiments provide a method for clustering a set of data compute nodes (DCNs), which communicate with each other more frequently, on one or more host machines. The method groups together guest DCNs (GDCNs) that (1) execute on different host machines and (2) exchange network data among themselves more frequently, in order to reduce interhost network traffic. The more frequently-communicating GDCNs can be a set of GDCNs that implement a distributed application, GDCNs of a particular tier in a multi-tier network architecture (e.g., a web tier in a three-tier architecture), GDCNs that are dedicated to a particular tenant in a hosting system, or any other set of GDCNs that exchange data among each other regularly for a particular purpose.

In a distributed cloud environment, a collecting agent deployed external to a kernel of a compute host collects network data packets describing various raw events communicated between compute instances of the distributed cloud environment and metadata associated with the events from various sources. The collecting agent communicates the metadata to a cloud service where it may be stored. The collecting agent communicates the packets to a stream processor that is decoupled from the collecting agent. The stream processor processes the packets in a stateful manner to generate a set of time series data. The time series data is communicated to the cloud application, where a set of enhanced time series data is generated by merging the time series data with the metadata in a distributed manner. A topology map describing the compute instances of the distributed cloud environment is then generated based on the set of enhanced time series data.

A method and system are provided to facilitate the commissioning of a distributed system. The method and system obtains a base configuration which defines an expected virtual topology for a distributed system, and identifies via a control device a physical network topology of the distributed system to commission the distributed system. The control device performs network discovery to identify a plurality of connected devices that are communicatively coupled thereto, and to collect device information for each connected device relating to its identity and relative position in the distributed system which has connected devices or associated subnetworks connected in a ring network topology. The collected device information for the plurality of connected devices is compared to the expected virtual topology from the base configuration to determine an identity and physical location of the plurality of connected devices and associated subnetworks in the physical network topology of the distributed system.