Just my view on Solarwinds DPA. Not worth it to me.

It focuses on tracking/reporting the wait time statistic as the core measurement for performance. I think to most DBAs and SQL Developers, wait time is a symptom of an issue, not the root cause. I would prefer this premise be changed to track different issues that get closer to the heart of the issue.

"Solutions" to wait time issues that the system suggests is simply canned responses by each wait type. This is incredibly unhelpful when tracking specific issues; for instance, a single query that is not performing well.

Navigating the UI is incredibly frustrating. Trying to click a wait bar and navigating to a SQL query by its hash is rather obnoxious as well. Hovering over every hash to find the query you might be looking for is just a pain.

Combining multiple search patterns is nearly impossible by any known means. For instance, I want to know information about queries on a database, by user, not by wait type. I have a page for queries against a database and I have a page for queries ran by a specific user. But I need ways to combine these quickly to find things.

Completely absent features like tracking SQL Agent jobs and maintenance plans would be a huge help.

An easier way to get to and use the "Current->Active Sessions." Currently running (or blocked) transactions are some of the most helpful tools I can have when determining issues. Focusing more on this feature would be a big plus.

Reports are halfway useless. It seems like they only give me the options to show what I can view in the UI already, instead of extending functionality. If I need to write every report I want with a SQL Query, then what did I need to buy this tool for in the first place?

More high-level information on the dashboard could take up some unused space. Things like: data file size, log size, agent jobs running/failed, etc could all be beneficial to show at a high level instead of a bunch of arbitrary green check marks.

One of the other big issues I have is when a blocking transaction goes unnoticed for a while. It takes up so much of the 'wait time' graphs that other bars are basically flat. I know there is a way to make it so that query is ignored in the wait bar graphs, but finding that option is a problem. And, I don't really want to ignore it all the time. I want to ignore the single huge chunk that was being blocked a week ago so I can continue monitoring for issues with it. The other thing is that when a blocking transaction occurs, a bunch of other queries rack up massive wait time too, which just compounds the problem.