The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
In the DCNM 10.4.2 release, the following two features have been introduced as previews for LAN deployments.
The concept of preview features is to get early feedback from customers for subsequent integration into the General Availability (GA) release. Note that preview features have not been qualified for any scaled environments.
Compute Visibility with VMware vSphere Virtual Center (vCenter)
Environmental Metrics visibility via streaming telemetry
Note | These features are only available in the OVA form factor for DCNM 10.4.2. The installer options in which these features are supported are (1) VXLAN Fabric (2) LAN, SAN, Auto-Config |
This section contains the following topics:
In virtualized environments, any kind of troubleshooting begins with identifying the network attachment point for the virtual machines in question. This means a quick determination of the server, virtual switch, port-group, VLAN, associated network switch and physical port is critical. This may require multiple touch points and interactions between the server and network administrator as well as reference to multiple tools (compute orchestrator, compute manager, network manager, network controller etc.).
The compute visibility preview feature added in DCNM 10.4.2 strives to provide a single pane of glass that provides the entire attachment path from the end-host all the way up to the network switch. The current DCNM network topology view has been enhanced to optionally display the associated compute nodes including virtualized servers, virtual switches, and virtual machines. Fabric level scope select filters the view to only display compute resources associated with the selected group of switches. This feature is targeted for VMware vSphere vCenter environments.
In today’s data center environments, granular visibility and tracking of network events has become extremely critical. The traditional polling based methods that pull network state in predefined intervals need a fork-lift upgrade. More advanced streaming approaches are required that provide network event visibility in closer to real time through a push method. Streaming telemetry not only allows data to be pushed out at a much finer granularity with a lower cadence (shorter interval) but it also enables event based notifications. While getting relevant data in a timely fashion is highly desirable, the data needs to be analyzed and converted into actionable insights.
As a first step toward LAN analytics, DCNM 10.4.2 enables subscriptions for environmental metrics through streaming telemetry for consumption and analysis. The environmental metrics that are streamed include CPU, Memory, Power, and Fan Speed; all of these are enabled with a single click. DCNM configures the streaming updates for CPU, Memory to be streamed out every 30 seconds, and those for Power, Fan Speed to be streamed out every 300 seconds (5 minutes). The per-metric real-time streaming dashboards allow filtering on a per fabric and per switch level including a per-switch drill-down where applicable. Streaming telemetry is currently supported on the Nexus 9000 platforms.
The pre-requisites for enabling this feature are:
The Cisco Nexus 9000 switches and Cisco DCNM need to be time synchronized (NTP is recommended).
Minimum software version on the Nexus 9000 switches must be 7.0(3)I6(1) or higher.
The memory dashboard depicts the actual memory consumption (RAM) on every switch in Gigabytes (GB). The per-process memory consumption will be available at a later stage.
The top-level view for the power dashboard depicts the efficiency of the power supplies. By definition, efficiency is Output-Power/Input-Power, which consequently results in a maximum efficiency of 1.0.
The drill down view indicates the input power in Watts, consumed by the individual power supply of the selected switch. A sample drill-down view for a selected device is shown below (n9k-112-spine):
The top-level view for the fan dashboard depicts the speed difference between the various fans in the system. In the typical case, the expectation is that all the fans in the same tray, operate at more or less the same speed.
The drill down view provides the fan speed as a percentage metric of each individual fan within a selected device. A sample drill-down view for a selected device is shown below (n9k-bg1):
The LAN Telemetry Health screen, provides a detailed break-down of how much data is being streamed out by each switch per feature for the last 24 hours. This data can be accessed by traversing the menu: Monitor > LAN Telemetry > Health.
The bar graph depicts the top 5 streaming switches and a has drill-down capability for feature-wise break-down.