Extended Events have been available since SQL Server 2008, but they are still a black box to most database administrators. Rather than focusing on the technicalities of sessions and targets, Jes will show the value of Extended Events by showing you what problems can be solved. Metrics can be gathered, performance problems can be targeted, and events can be tracked. Let me show you how easy it is!
Brent Says: The timeline in here showing XE changes from SQL Server 2008 to SQL 2014 is really helpful. Profiler is a dead man walking, and XE is more like a zombie that rose up from the dead and became a powerful superhero working for the good guys. (Look, writing analogies isn’t my day job.)
Kendra Says: I agree– Profiler stinks! I especially love Jes’ analysis of what SQL Trace had that we don’t have in Extended Events: super helpful and not something you’d notice right away. I’m totally on Team Extended Events, I just like knowing the little things I may need to code or manage around.
7 Comments. Leave new
Thanks Jes!
You have sold me on the idea of looking into it. I have been a perfmon/profiler guy.
Will you actually be demonstrating the tool in a future session?
Thanks!
I will be! I’ve been working with it a lot in the last couple of months. You can check out one video on querying the targets for an example!
Jes, thanks for this!
You make a great case for Extended Events. I will definitely be investigating further, and staying tuned for more great videos from you and your team.
Glad to hear it Brett!
What I think will be cool for extended events would be a, 911-EX. (like the first Responder’s kit you have for download, a similar, but thru Extended Events.)
Ranga – can you elaborate more on what you mean?
Brent,
I meant setting up a bunch of extended events to monitor a sql server which is performing very badly and call it 911-XE . For example queries running for 5 seconds or more ( 2 is normal, but, we are trying to catch something abnormal, like someone running a report query ), perfmon IO stats breaching a threshold, PLE dropping below a threshold, spids that are consuming high resources etc etc…something like when a paramedic reaches an accident location and gathers bunch of parameters like pulse, BP etc and then takes the next course of action….the idea here is to run this XE session for few minutes and then analyse the info to find the root cause for the server’s poor performance…we can always change the threshold numbers as needed since each environment might be different…this XE sessions can be used when the DBA gets a call like “Our application is very slow since last 20 minutes…queries are timing out, we need HELP”, you should have got this call few times as a DBA, haven’t you 🙂