The Flow Insight page provides an insight into data centers, devices, and flows. It is a context-based page as it performs analysis based on the entities, flows, and the time range that you select.

To access the Flow Insight page, see Access Flow Insight page.

Alternatively, you can search for Flows and in the search result page, click Flow Insight.

The various sections in the Flow Analytics Dashboard are:
  • Top Talkers
  • What's New
  • Network Performance
  • Outliers

Top Talkers

This section helps you to recognize which entities are talking the most in your environment. You can select different kinds of entities such as Source-Destination pair, VM, Cluster, L2 Network, Subnet. This widget lists the top 10 talkers in the entity category that you select. It helps the customer to plan for network optimization. The metrics that are represented by bars in this widget are as follows:
  • By Flow Volume: Indicates the traffic volume.
  • By Traffic Rate: Indicates the rate of traffic.
  • By Session Count: Indicates the number of sessions.
  • By Flow Count: Indicates the number of flows
Note:
  • If a VM appears in one or more metrics, when you point to that VM in a bar, it will also be highlighted in other bars.
  • When you click a VM in the metrics bar, the complete list of flows coming to this VM is shown.
  • When you select VM as the entity in the Top Talkers list, all the flows related to this VM irrespective of it being the source or destination is shown. If you select Source VM in the list, then only the flows coming from this VM are considered.
  • If you are considering the physical flows, you can select either Source IP or Destination IP.
  • After you select the Source-Destination pair and point on the metric bar, if you click the link in the tool tip, the corresponding dashboard appears. For example, for a VM in Sourec-Destination pair, the VM-VM path dashboard appears.
  • For a flow group view or a flow entity projection or a flows group query, you cannot see the Flow Analytics button.

What's New

This section helps you to track what services and entities are discovered in the data center in the selected time range. The widget in this section are as follows:
  • New Virtual Machines Accessing Internet: Lists the new VMs that access Internet.
  • New Internet Services Accessed: Lists the new Internet services discovered in the environment.
  • New Internal Services Accessed: Lists the new intranet services that are discovered and accessed from the Internet endpoint.
  • New Internal/E-W Services Accessed: Lists the services that are exposed and accessed by the machines within a data center
  • New Services with Blocked Flows: Lists services that have blocked flows. This section is populated only for IPFIX.
  • New Firewall Rule Hits: Lists the new firewall rules that are brought into effect. This section is populated only for IPFIX.
  • New App-Ids: Lists the new application IDs. This section is populated only for NSX Intelligence.

Network Performance

In this section, you can find and visualize the abnormal flows for the various ranges of TCP Round Trip Time (RTT) values based on the selected criteria.
Note: vRealize Network Insight Cloud shows average TCP RTT metrics at 5 minutes granularity for last 24 hours only.
If the flow deviation percentage is 100 percent and absolute deviation is 20 milliseconds (ms), then vRealize Network Insight Cloud considers that flow as an abnormal flow.
In the visualization, the left side shows the different range of TCP RTT and the right sides shows the normal and abnormal deviation range. Based on the values of percentage deviation and absolute deviation, flows are connected from left (TCP RTT) to right (DEVIATION). You can analyze the following types of flows:
  • Inter-Host
  • Intra-Host
  • Internet
  • All Flows
You can also change the percent deviation and absolute deviation based on your requirement.

In the following example, there are two different ranges of TCP RTT, one is less than equal to 30 ms and the other is 30-120 ms. You can find there are total 151 flows comes under the less than equal to 30 ms TCP RTT range. Out of the 151 flows, 9 flows are shown as an abnormal flow.



To get more insight about the TCP RTT distribution information and counts of flows, click the colored line in the visualization. In the following example, you can see the detail insight about the TCP RTT distribution information and counts of flows:



Outliers

This section helps you to track and analyze related data. It consists of the following sections:
  • Elephant Flows: This section helps to identify the flows which have small count of sessions and high throughput versus flows which have large count of sessions and small throughput. Typically, the flows with the large session counts and small throughput are also referred as mice flows. The analysis is based on the ratio of bytes to the number of sessions. Each dot in the graph represents multiple flows. When you point to a dot, you can see the list of flows. To view the details of a particular flow, click that flow in the list.
  • Custom Analysis: This section allows you to visualize the flow data on 2 dimensions of your choice. It helps in analyzing the data to find the outliers in various ways.
    Note: The metrics represented in this section are the approximate values and not the exact values.