Abstract: An example method comprises determining, by an edge services controller, based on a respective predicted resource utilization value for each of a plurality of servers, a corresponding server weight for each of the plurality of servers; the plurality of servers comprising respective network interface cards (NICs), wherein each NIC of the plurality of NICs comprises an embedded switch and a processing unit coupled to the embedded switch; determining, by the edge services controller, based on a respective predicted resource utilization value for each of a plurality of services, a corresponding application weight for each of the plurality of services; and scheduling, by the edge services controller, based on the respective server weight for a server of the plurality of servers and the respective application weight for the service, a service of the plurality of services on the server.

Abstract: A system comprises an edge services controller configured to: compute, based on a physical topology of physical links that connect a plurality of network interface cards (NICs) that comprise embedded switches and processing units coupled to the embedded switches, a virtual topology comprising a strict subset of the physical links; and program the virtual topology into the respective processing units of the NICs to cause the processing units of the NICs to send data packets via physical links in the strict subset of the physical links.

Abstract: In general, techniques are described for automatically and transparently providing service proxying to virtual machines using Network Interface Cards (NICs). In some examples, a service proxy implemented by a NIC of a computing device that hosts a virtual machine may mimic sidecar service proxy behavior. A NIC-based mesh agent may automatically detect the service offered by the virtual machine and interact with control plane components to dynamically incorporate the service into a service mesh.

Abstract: An example apparatus to manage network resources includes a link aggregator to: aggregate a first plurality of physical network interface cards to create a first link aggregated group, the first link aggregated group corresponding to a first virtual network interface card; and aggregate a second plurality of physical network interface cards to create a second link aggregated group, the second link aggregated group corresponding to a second virtual network interface card; and a link manager to: connect the first link aggregated group between a first distributed virtual port and a first top-of-rack switch; connect the second link aggregated group between a second distributed virtual port and a second top-of-rack switch; and remove an invalid connection between the first plurality of physical network interface cards and the second top-of-rack switch.

Abstract: A computing system comprising a memory and processing circuitry may perform the techniques. The memory may store time series data comprising measurements of one or more performance indicators. The processing circuitry may determine, based on the time series data, an anomaly in the performance of the network system, and create, based on the time series data, a knowledge graph. The processing circuitry may determine, in response to detecting the anomaly, and based on the knowledge graph and a machine learning (ML) model trained with previous time series data, a causality graph. The processing circuitry may determine a weighting for each edge in the causality graph, determine, based on the edges in the causality graph, a candidate root cause associated with the anomalies, and determine a ranking of the candidate root cause based on the weighting. The analysis framework system may output at least a portion of the ranking.

Abstract: An example system including a first top-of-rack (ToR) switch, and a second ToR switch, the second ToR switch is to receive a network packet from a first host, the network packet to include a destination address of a second host, and after a failure of a physical network interface card (pNIC) at the second host eliminates a first link between the second host and the second ToR switch, send the network packet to the first ToR switch via an inter-switch link between the first and second ToR switches to cause the first ToR switch to forward the network packet to the second host via a second link between the first ToR switch and the second host.

Abstract: A disclosed example includes configuring a virtual network and a physical network for at least twenty-four hosts and a top-of-rack switch in less than one minute by: generating, by executing an instruction with one or more processors, network topologies of the hosts based on physical network connection information indicative of physical network connections between the hosts and the top-of-rack switch; determining, by executing an instruction with the one or more processors, whether implementing the network topologies of the hosts concurrently is valid based on an evaluation of the network topologies relative to a network topology validation rule; and configuring, by executing an instruction with the one or more processors, the top-of-rack switch based on one or more of the network topologies.

Abstract: Techniques are described for determining the energy usage of a data center and invoking one or more actions to improve the energy usage of the data center. For example, a computing system may obtain energy usage data of a data center. The computing system may also determine, based on a comparison of the energy usage data of the data center to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the data center that specifies a value that indicates whether the data center is energy efficient. The computing system may further invoke, based on the green quotient of the data center that specifies a value that indicates the data center is not energy efficient, an action to improve energy usage of the data center.

Abstract: Techniques are described for determining the energy usage of a data center and invoking one or more actions to improve the energy usage of the data center. For example, a computing system may obtain energy usage data of a data center deploying an application. The computing system may also determine, based on a comparison of the energy usage data of the data center deploying the application to a percentage of energy provided by one or more renewable energy sources to the data center, a green quotient of the application that specifies a value that indicates whether the data center deploying the application is energy efficient. The computing system may further invoke, based on the green quotient of the application that specifies a value that indicates the data center deploying the application is not energy efficient, an action to improve energy usage of the data center deploying the application.

Abstract: A performance monitoring system includes a metric collector configured to receive, via metric exporters, telemetry data comprising metrics related to a network of computing devices. A metric time series database stores related metrics. An alert rule evaluator service is configured to evaluate rules using stored metrics. The performance monitoring system may include a machine learning module and is configured to determine optimized metric collection sampling intervals and rule evaluation intervals, and to automatically determine recommended alert rules.

Abstract: A performance monitoring system includes a metric collector configured to receive, via metric exporters, telemetry data comprising metrics related to a network of computing devices. A metric time series database stores related metrics. An alert rule evaluator service is configured to evaluate rules using stored metrics. The performance monitoring system may include a machine learning module and is configured to automatically determine recommended alert rules.

Abstract: A performance monitoring system includes a metric collector configured to receive, via metric exporters, telemetry data comprising metrics related to a network of computing devices. A metric time series database stores related metrics. An alert rule evaluator service is configured to evaluate rules using stored metrics. The performance monitoring system may include a machine learning module and is configured to determine optimized metric collection sampling intervals and rule evaluation intervals, and to automatically determine recommended alert rules.

Abstract: The disclosure herein describes a virtual extensible local area network (VXLAN) gateway. During operation, the VXLAN gateway receives, from a physical host, an Ethernet packet destined for a virtual machine residing in a remote layer-2 network broadcast domain that is different from a local layer-2 network broadcast domain where the physical host resides. The VXLAN gateway then determines a VXLAN identifier for the received Ethernet packet. The VXLAN gateway further encapsulates the Ethernet packet with the virtual extensible local area network identifier and an Internet Protocol (IP) header, and forwards the encapsulated packet to an IP network, thereby allowing the packet to be transported to the virtual machine via the IP network and allowing the remote layer-2 network broadcast domain and the local layer-2 network broadcast domain to be part of a common layer-2 broadcast domain.

Abstract: An example first server host includes processor circuitry to: connect a virtual network interface card (vNIC) of the first server host to a first physical network interface card (pNIC) and a second pNIC of the first server host; establish an inter-switch link between first and second switches, the first switch and the first server host in a first network fabric, the second switch and a second server host in a second network fabric; and cause transmission of a first and second network packets from the vNIC of the first server host, the first and second network packets to be delivered to the second server host via the inter-switch link, the first network packet to be transmitted via the first pNIC when utilization of the first pNIC does not satisfy a threshold, the second network packet to be transmitted via the second pNIC when the utilization satisfies the threshold.

Abstract: The technology disclosed herein enables an L3 network fabric including one or more spine switches having a leaf-spine topology to be self-expanded. In a particular embodiment, a method provides transferring one or more probe messages from each of the spine switches. The probe messages detect whether new computing nodes have been attached to the communication network. The method further provides receiving a reply to at least one of the probe messages. The reply identifies a new computing node that is not yet included in the L3 fabric.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to improve workload domain management of virtualized server systems. An example system includes memory, programmable circuitry, and instructions to program the programmable circuitry to generate a pool of virtualized servers based on a policy, determine whether a utilization of a first virtualized server is less than a first threshold, the first threshold based on at least one type of resource provisionable to the first virtualized server, determine that a firmware status associated with the first virtualized server corresponds to a first firmware version, transfer a workload of the first virtualized server to a second virtualized server after a determination that a second firmware version is available for the first virtualized server, deallocate the first virtualized server from the first workload domain to the pool of the virtualized servers, and update the first virtualized server to the second firmware version.

Abstract: Some embodiments of the invention provide a novel method of performing network slice-based operations on a data message at a hardware forwarding element (HFE) in a network. For a received data message flow, the method has the HFE identify a network slice associated with the received data message flow. This network slice in some embodiments is associated with a set of operations to be performed on the data message by several network elements, including one or more machines executing on one or more computers in the network. Once the network slice is identified, the method has the HFE process the data message flow based on a rule that applies to data messages associated with the identified slice.

Abstract: A container orchestration platform manages a plurality of instances of resources including a first custom resource and a second custom resource. An API server of the container orchestration platform receives a request to delete an instance of the second custom resource; determines whether instance data associated with the instance of the second custom resource has a backreference identifying an instance of the first custom resource, the backreference indicating the instance of the first custom resource is dependent on the instance of the second custom resource; and in response to determining that the instance data has the backreference to the instance of the first custom resource, bypasses deletion of the instance of the second custom resource.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to improve workload domain management of virtualized server systems. An example apparatus includes a resource pool handler to generate a pool of virtualized servers including a first virtualized server based on a policy, ones of the virtualized servers to be allocated to a workload domain to execute an application, a resource status analyzer to determine a health status associated with the workload domain and determine whether the health status satisfies a threshold based on the policy, and a resource allocator to allocate the first virtualized server to the workload domain to execute the application when the health status is determined to satisfy the threshold.

Abstract: A disclosed example includes: a resource utilization analyzer to determine 1) first workloads of a first workload type deployed in a first server room in a data center, and 2) second workloads of a second workload type deployed in the first server room; a workload authorizer to determine that first virtual machines executing the first workloads and second virtual machines executing the second workloads cause a first server rack to generate an amount of heat; and a migrator to migrate the first virtual machines from the first server rack of the first server room to a second server rack of a second server room in the data center to reduce a temperature in the first server room based on the amount of heat, the migrator to migrate the first virtual machines to the second server rack without migrating the second virtual machines to the second server rack.