A Simple Alert aggregates over all reporting sources.
You will get one alert when the aggregated value meets the set conditions. This works best to monitor a metric from a single host, like
max of system.cpu.iowait over host:xxx or for an aggregate metric across many hosts, like
sum of nginx.net.request_per_s over availability-zone:us-east-1b
A Multi Alert applies the alert to each source, according to your group parameters.
To alert on disk space you might group by
device, creating the query:
avg of system.disk.in_use over name:cassandra by host,device. This will trigger a separate alert for each device on each host tagged
name:cassandra when the alert conditions are met. Note: Anything reporting this metric that does not have the chosen groups will be ignored during alert evaluation.
Your Multi-alert monitor will automatically identify new scopes to monitor over within the tag groups you've chosen, which makes the multi-alert setup especially useful for monitoring on environments that quickly or automatically scale up/down.
It takes a multi-alert monitor approximately 1 minute to identify a new tag group when it becomes available for monitoring.
When a tag group stops reporting data, the monitor will "drop" that tag group after it reports no-data for a consecutive 24 hours.