I know this is very late down the road, so probably no help to you, but maybe for others that read:
Dynatrace F5 Pros:
- Polls every 1 minute versus SolarWinds polling every 10(hosts)/9(interfaces) minutes (by default. You can set SolarWinds to poll more often, but their polling engines are not really designed for every 1-minute polling and may buckle under the pressure if you have a lot of devices or really big F5 devices)
- Can connect Pool Members to the Process Instances that are behind them. So not only can it map Pool Members to the hosts, but it can map them to the individual processes on the host, like the IIS or Apache Tomcat web process for example. This is HUGE.
- In Distributed Traces, Dynatrace can tell you which F5 Virtual Server or Pool a specific trace is proxied through. Dynatrace calls it "Proxy Chaining". It can be super helpful for troubleshooting with traces and knowing which middleware layers a trace is traversing.
- Grail! Grail is awesome and having the metrics and entity metadata in Grail allows you to build incredible queries, dashboards, and alerts with all the fantastic and enriched data.
- Easily customizable charts on the go. In SolarWinds, you have to have the ability to edit Views and Widgets to be able to change the time range on charts. And each chart has its own time range. With Dynatrace's time range picker that updates all charts on a dashboard, getting to the data you need is super easy. The only way for an average user to do this in SolarWinds is with their Perfstack charts which are not super easy to get to and configure for an average user. Dynatrace wins here by a pretty huge amount.
Dynatrace F5 Cons:
- Dynatrace is not a network tool, so some of the metrics aren't quite as accurate as what you see in SolarWinds. For example, the interface bytes in/out don't report in Bytes/Bits per second. Yes, you can adjust the metrics to report that way, but the values never seem to add up accurately when doing that. Also, Dynatrace doesn't natively collect the speed of the interface and calculate percent utilization like SolarWinds does. Again, you can do this with metric formulas, but you have to know what the speed of an interface is and do the metrics yourself. It's a pretty big pain and a big weak point in my opinion.
- For really big F5 instances (ones that have a lot of entities and traffic on them), the SNMP jobs Dynatrace runs can really fill up the management interfaces on the F5 devices. Especially considering a lot of F5 devices (even newer ones) only have 100Mbps management interfaces (which is bananas. It's 2025 F5. 1Gbps interfaces should have been the minimum interface speed for anything you've made in the last 10 or 15 years, even management interfaces!).
- SolarWinds F5 pages have cool little visualization maps that show the device/virtual server/pool/members dependency and their status. Dynatrace really needs to improve the map visualization capabilities in general (SmartScape is pretty awful right now. I can't wait for the new SmartScape!)
- Drill-down capabilities. SolarWinds is getting old, but one thing they do well in general is drill-down, drill-down, drill-down. Pretty much everything on any SolarWinds dashboard is clickable and takes you right to the item you're clicking on. No dots to locate and click and then hit Open With and then choose an app (if one is even available). Nope. You click; you go right to it. It makes navigating and troubleshooting extremely easy in SolarWinds. The Dynatrace dashboard team desperately needs to improve their drill-down capabilities. They have this new Dashboards 3.0 framework, and they still didn't really fix that issue (they made it a bit better, but it is still very far away from a tool like SolarWinds in this regard).
