Monitoring Redis Enterprise Cloud Performance
Redis Enterprise Cloud (RC) provides a straightforward dashboard that gives you good visibility into each database. Metrics can be viewed on the Metrics tab of a selected database. Use the scroll bar at the top to change between time intervals: last minute, five minutes, hour, day, week, month, and year.
If you hover your cursor over a metric's mini-graph, you should see two boxes with arrows, one arrow pointing to the left and one to the right (depending on which side of the mini-graph you hover on). Each side of the mini-graph corresponds to the larger moving graphs towards the top of the page, under the timescale For example, if in the big graphs you want to see Reads/sec on the left and Writes/sec on the right, click on the left side of the Read/sec mini-graph and the right side of the Writes/sec mini-graph.
For a quick tour of what you get, watch this video.
Definition of Each Metric
For each database, there are 14 different metrics you can see on the page:
|Ops/sec||The number of overall operations per sec for all Redis commands|
|Reads/sec||The number of read operations per second|
|Writes/sec||The number of write operations per second|
|Other cmds/sec||The number of other Redis commands per second|
|Latency (in milliseconds)||Latency per write operation|
|Reads Latency (in milliseconds)||The average, min, max and last values are also shown|
|Writes Latency (in milliseconds)||Latency per write operation|
|Other Latency (in milliseconds)||Latency per other commands|
|Used Memory||The amount of memory used by the database|
|Total Keys||The total number of keys in the database|
|Connections||The total number of connections to the end point|
|Evicted Objects/sec||Number of objects evicted from the database per second|
|Expired Objects/sec||Number of expired objects per sec. An expired object is an object with expired TTL that was deleted from the database.|
|Hit Ratio (percentage)||The number of operations on existing keys divided by total database operations.|
Alerting on Metrics
There are five metrics, located on the Configuration tab for each database, that you can and should create alerts for.
If you want to edit the values or enable/disable an alert, you must first click on the pencil icon to edit the configuration. While there are default values, each database's alerting needs are usually different, so there are no best practices, per se, around what to set these to.
Note: The "Total size of datasets under this plan" metric is at the subscription plan level even though they are shown (for convenience) at the database level. If you set this alert in one database you will see the value has changed in the other database configurations too. The exception to that is Pay-as-you-go plans. For "Number of connections reached," each database can have its own threshold. If you have multiple databases on the same subscription, you need to set this percentage appropriately on your own. For example, if you have two databases, configuring an alert for when each database reaches 80% can surpass your number of subscriptions.
Changing Alert Email Recipients
Any member of the account team can receive alert emails.
To change the alert email recipients please do the following:
- Click the 'Team' button in the top right of the main UI page.
- Click 'Edit'.
- Check the 'Receive Email Alerts' checkbox of the users who should receive alerts.
- Save your changes by clicking 'Save'.
NOTE: PaaS users, such as Heroku users, don't have the option to create a team. Users on a team can specify an alert email and a billing email, regardless of what email is set in Account -> Account Info. For PaaS users, on the other hand, the alert emails and billing emails are sent to the email that is set in Account -> Account Info.