TECHNICAL DATA

SDN Monitoring

Entuity Network Analytics provides support for both Cisco ACI and VMware NSX software-defined networking technologies.

CISCO ACI MONITORING

Cisco’s ACI (Application Centric Infrastructure) has been developed to simplify and ease the transition to SDN to meet demands for agile methodologies such as DevOps. ENA matches this demand with comprehensive support for Cisco ACI that extends network management beyond SNMP to include tenants, pods, application profiles and endpoint groups, etc.

ENA monitors the Cisco Nexus 9000 Series Switches on which the APIC (Application Policy Infrastructure Controller) operates. We can take the APIC under management, show you how it’s configured, and alert you to issues that arise on it. Correlating data on the Nexus switches and the APIC itself, we give you comprehensive visibility into the running of your APIC and Nexus kit.

Discover Cisco ACI deployments controlled by the APIC

Once you have taken the Cisco APIC under management, ENA will immediately detect any new Cisco ACI deployments controlled by that APIC and then begin monitoring them. Extend your network whenever and wherever you want, and ENA will keep an eye on it for you.

Measure 10 metrics from the APIC

Gain insight into every part of the Cisco APIC, which is the central component of the ACI. ENA measures the following metrics, covering all aspects of the APIC model:

R
APIC CPU Utilization %
R

APIC Memory Utilization %

R

Endpoint Group Health

R

Fabric Node Health

R

Fan Tray Health

R

Line Card Health

R

Pod Health

R

Power Supply Health

R

Supervisor Card Health

R

Tenant Health

Maintain the health of your Cisco APIC

For all metrics (except APIC CPU and APIC Memory), APIC health is measured as a percentage.

  • 100% = healthy.
  • 0% = unhealthy.

Cisco advises that if health drops below 90%, then there is a problem. Therefore, ENA sets 90% as the default health threshold. When a metric’s health drops below 90%, ENA generates an event on that metric.

For the APIC CPU and APIC Memory metrics, ENA uses a percentage based on usage:

  • 100% usage = unhealthy.

 

Reduce MTTRS with ACI-specific events and incidents

Entuity Network Analytics delivers a whole range of APIC-specific events and incidents to help you get straight to the cause of any issues that might arise, fix them, and resume service.

 

  • SDN Controller Poll Failure.
  • SDN Controller Poll Failure Cleared.
  • ACI Low APIC Memory (relative to threshold).
  • ACI Low APIC Memory Cleared.
  • ACI Low APIC CPU (relative to threshold).
  • ACI High APIC CPU Cleared.
  • ACI Low EPG Health (relative to threshold).
  • ACI Low EPG Health Cleared.
  • ACI Low Tenant Health (relative to threshold).
  • ACI Low Tenant Health Cleared.
  • ACI Low Supervisor Card Health (relative to threshold).
  • ACI Low Supervisor Card Health Cleared.
  • ACI Low PSU Health (relative to threshold).
  • ACI Low PSU Health Cleared.
  • ACI Low Fan Tray Health (relative to threshold).
  • ACI Low Fan Tray Health Cleared.
  • ACI Low Line Card Health (relative to threshold).
  • ACI Low Line Card Health Cleared.
  • ACI Low Fabric Health (relative to threshold).
  • ACI Low Fabric Health Cleared.
  • ACI Low Pod Health (relative to threshold).
  • ACI Low Pod Health Cleared.

 

Correlate ACI faults with wider network events and incidents

ENA also measures the ACI faults on Cisco APIC components, giving you an extra layer of insight into the health of your SDN and the causes of health-related Cisco APIC events and incidents.

If ENA raises events and incidents against a particular Cisco APIC health metric, use the ACI Fault dashlet to view the ACI faults on that component and quickly see exactly what is impacting the APIC.

ACI faults are a separate concept to ENA events and incidents, and only exist within the APIC itself.

Meet the challenge of scaling networks

Cisco ACI deployments are all about increasing the efficiency of network provisioning and making it easier to scale. ENA is endlessly scalable without suffering performance issues, making it the perfect solution for growing businesses and service providers taking on more customers. Keep on top of your increasingly dynamic and distributed network, from legacy installations to Cisco ACI’s SDN deployments, in one solution.

Monitor SDN as part of your wider infrastructure

Network components don’t exist in isolation. You need the ability to see the bigger picture. ENA’s all-in-one, dashboard-centric UI enables you to:

View your Cisco APIC performance in as much or little granularity as you want through specific summary and ACI fault dashboards. Drill down into ACI pods and see all their subordinate components and their relationships, as well as viewing their statistics and configurations. Browse dashboards specific to the APIC hardware and tenants that make up your ACI, providing all of the information necessary to ensure it delivers the network service your users need.

See how your ACI fits in with the rest of your network through any number of our flexible and customizable dynamic dashboards, meaning you can monitor issues and find their causes with minimal effort.

Directly manage the switches in the APIC by SNMP, allowing you to draw a whole host of complementary information, such as topology, port-level activity, resource utilization and flow export capability.

Understand how your Cisco ACI deployment is affecting the business services your network provides through ENA’s service functionality.

Manage your Cisco ACI in a way that best suits you.

Future developments

Entuity is currently developing functionality to enable discovery and monitoring of the elements attached to the ACI fabric. This will give you insight into the objects such as hypervisors, virtual machines and physical servers running on the leaves of the APIC. Complete your network monitoring visibility with analytics of the physical and virtual elements across your legacy and software-defined network.

VMware NSX VXLAN SUPPORT

ENA supports monitoring of VMware NSX SDN flows (i.e. VXLAN overlays). Fully integrated with and utilizing ENA’s wider functionality, our VXLAN SDN monitoring gives you detailed statistics for both native IP flows as well as NSX VXLAN-encapsulated traffic.

Augment flow reporting

Further enhance your flow reporting with ENA’s NSX VXLAN coverage, giving you details about your VM endpoints so that you can see the VXLAN-only breakdowns of flow on your VMware NSX devices, and report on true VM to VM conversation statistics.

Quickly find causes and impacts of heavy link traffic

You might have a link in the network or cloud between a pair of routers that is maxed out, or you want to take it down for maintenance. How can you tell what this will impact?

ENA unifies VXLAN and flow monitoring to give you insight into the flows on your virtual distributed switches, so you can find out which virtual machine is using which link in your network or cloud. Filter the appropriate IP addresses to find which VMs are talking to each other and traffic volumes, for example. Discover why links are hot, who’s using specific links, and if you can safely take a link down.

Monitor the whole network

Automated discovery of VMs alongside the other devices and components of your network puts your SDN in the context of its environment. Understand traffic rates, packet rates, application breakdowns and QoS settings, so that you can help provide optimal performance across both physical and virtual networks.

Entuity proactively highlighted areas to address before they became service-impacting issues, allowing us to ultimately provide better services to our customers.

 

Kristin Morris

Manager, Network Management Applications, Dell

It is vital that our network is running to the best of its availability at all times. Entuity has enabled us to make effective use of capacity, as well as identify faults before they have had any significant impact.

 

Manager of Network Services

Large Financial Services Company

Entuity satisfied our network management system requirements with its all-in-one solution—minimizing the number of tools we need, helping us manage our costs and deliver services more effectively.

 

Mike Faust

Network Design Engineer, University of Minnesota