The Defined Time field at the top left of the Data Flow page enables you to select the sampling time for which to display data flow analytics. The options available are:
Real time. This is the default. This queries the last sample from within the last three minutes, which refreshes automatically (unless you switch auto refresh off). To select real time, click once, select Select Time Range and then select Real Time.
Predefined times. These options show data flow information over the selected time amount of time in the immediate past. To select a predefine time, click Defined Time, select Select Time Range and then select one of the following:
10 m. Show data flow sampled over the last ten minutes.
2 H. Show data flow sampled over the last two hours.
4 H. Show data flow sampled over the last four hours.
8 H. Show data flow sampled over the last eight hours.
1 D. Show data flow sampled over the last day.
3 D. Show data flow sampled over the last three days.
With these options, refreshing updates the data for the immediate past.
A custom time range. The analytics are averaged over the selected time range.
To select a custom time range:
Click Defined Time.
Select Select manually.
Click inside the Start field to set the start time.
Select a date and time from the dropdown box and click Apply inside the dropdown box.
Click inside the End field and repeat step 4 to set the end time.
Click the Apply button.
The nodes on the data flow paths (users, hosts, VIPs and so on) are sorted in their columns according to a combination of criteria. The page displays only the highest performers according to those criteria and according to a protocol filter, up to a limit set by the Limit field.
For any given column, you can click the sorting arrow toggle button () to switch the sorting order between highest to lowest and lowest to highest.
To change the global performance and protocol criteria, which determines sorting order as well as which nodes are included:
In the Sort by dropdown you can select one of the following:
BW (MB/S) (default). Sorts the objects in each column in order of average bandwidth over the sampling period.
IOPS. Sorts the objects in each column in order of average IOs per second over the sampling period.
MD IOPS. Sorts the objects in each column in order of average metadata IOs per second over over the sampling period.
In the Type of dropdown, you can select one of the following:
Total (default). The object with the most total IOs, within the selected Sort by criteria, is placed highest in the sorting order. For example, if IOPS is selected in Sort by, then the nodes with the highest average total IOPS are displayed as the highest performers.
Read. The object with the most read IOs, within the selected Sort by criteria, is the highest in the sorting order. For example, if IOPS is selected in Sort by, then the nodes with the highest average read IOPS are displayed as the highest performers.
Write. The object with the most write IOs, within the selected Sort by criteria, is the highest in the sorting order. For example, if IOPS is selected in Sort by, then the nodes with the highest average write IOPS are displayed as the highest performers.
In the Protocol protocol dropdown list, you can filter the display by access protocol:
All. Shows all IOs from all protocols.
NFS3. Shows only IOs sent via the NFSv3 protocol.
SMB2. Show only IOs sent via the SMBv2 protocol.
NFS4. Show only IOs sent via the NFSv4.1 protocol.
S3. Show only IOs sent via the S3 protocol.
A configurable limit applies to the number of nodes that can appear in each of the columns on the page.
Sorting criteria and filters set the criteria that determine which nodes are the top performers and appear on the page. Other nodes and the paths through those nodes are not displayed. They can be found by search.
To set this limit, select a value from the Limit dropdown. The default is 10.
When the number of nodes in more than one category is greater than the maximum, the number of nodes shown might sometimes be further reduced from the max limit value in each category. This is because nodes are only displayed if they share paths with the other displayed nodes.
In the Vip column you can use the slider to toggle between displaying the flow through each VIP individually and displaying the flow through each VIP pool.
Hover over any object on the page to see the IO metrics for the node, averaged over the sampled period:
The metrics displayed when you hover over an object reflect the aggregated flows included based on the sample period and the selected filters. They are therefore recalculated when change the sampling period, sort the nodes or change the limit. However, they are not recalculated when the node is pinned or the paths through the node are isolated as described in Focusing on Specific Flows and Nodes.
The search feature searches the sampled IOs for the selected time period, and can find a node and the IO paths that flowed through that node during the sampling period even if the node is a low performer according to the currently selected sorting and filtering criteria and doesn't appear due to the limit.
If searching doesn't find a given entity (such as a CNode, VIP, VIP pool or View), this indicates that IO didn't flow through it.
To search, enter text into the search field at the top of a column of nodes. Partial name text is valid and can return multiple results. For example, this search for "CNode-6" displays all the CNodes with names that include that string and the paths through those nodes:
The data reported by the Data Flow tool can be refreshed on demand or automatically as follows:
Automatic refresh is available when you are viewing real time data. It is on by default and you can switch it off and on again:
Manual refresh is available for all time selections. It refreshes data and recalculates the display. To refresh, click the manual refresh button:
Pinning nodes selects the data flow paths that go through those nodes. Those paths are shown more prominently than flows that do not go through the pinned objects. You can pin one or more nodes in one column at one time.
To pin a node, click on it once. To unpin a node, click it again.
When nodes are pinned, you can isolate all the flows that go through those objects. Isolating flows gives you a simplified view of the paths you want to focus on, while hiding all the paths that do not flow through the pinned nodes.
To isolate flows, first pin one or more nodes and then click Isolate Selected Flows. When you're done looking at the isolated flows, you can clear them by clicking Clear isolation.
Node pinning and isolated flows are features designed to aid you visually, enabling you to focus on a restricted set of flows. They do not perform any recalculation of the IO flows reflected on the page. Therefore, when you hover over a pinned node, the metrics displayed are the same metrics displayed before pinning and isolating. They reflect the aggregate of all the flows in the selected sampling period included by the filtering and sorting criteria.