brianflynn

I just browsed through reports available online and here's ones, that as a DBA, I would consider. Here's a couple of reports for blocking : https://thwack.solarwinds.com/docs/DOC-174921 https://thwack.solarwinds.com/docs/DOC-173995 Here's a report that can help you find table scans :…

mos, It seems that we can't currently do what you're asking so I've created a feature request and referenced this conversation. http://thwack.solarwinds.com/ideas/3653

Just to clarify, the SSAS template I've suggested you consider does not work with DPA. It works with SAM (Server Application Monitor). DPA does not presently support SSAS. aLTeReGo, perhaps you have guidance to share with loken.bhatt regarding implementing the SSAS template?

nicolas diogo, I'm not really clear on your question. It does seem that you are talking about Database Performance Analyzer (formerly Confio Ignite). DPA has an optional feature that requires an additional license. This feature enables DPA to juxtapose database performance data with VMware events so you can more easily…

OK, I understand. Well from my perspective, the primary advantage of SRM Orion, is the AppStack. Our goal is to have such a seamless integration between products that they behave as one product to solve real world, multi-silo problems. For example, being able to quickly identify an Application performing poorly is doing so…

So there should be 36GB free for the OS. What other things are running on the server? What does Task Manager say? Note, you can't trust Task Manager for SQL Server memory utilization in large memory configurations like yours. But you can see what other processes are using memory. Assume SQL Server is using all that you let…

deannich, that is a very clever solution. Please post back to let us know how it pans out. I'll bet other DPA users may like to implement this as well.

It is not yet available in SRM Orion module but is available in SRM Profiler module, of course.

Awesome! Glad I could help.

I'm happy to see good answers already posted here. Jim Thompson is right. Multiple threads executing in parallel is the answer. As far as what is happening, I see a locking wait type which jives with the query you shared because it is using a query hint demanding locks. That query hint tells the database instance to use…

As far as highlighting DPA 9.1 features which are also in 9.2, don't forget support for Amazon RDS. If you are beginning to work with cloud databases, this may be of interest. 

I see. There's a nuance that must not be clear to you. The bars don't represent all of the wait time. They represent wait time for the top X queries with the most significant wait time for that time period. In other words, your biggest targets for optimization. The wait for the other queries not shown do not get swatches…

Yes. The one customer I had in mind was doing that. They had DPA and an Orion front ends (probably SAM) installed on the same server.

OK. They're based on industry "best practices" which in my experience should be called "interesting opinions". :-) Often times, these best practice tips are based on something said by an expert when they were backed into a corner to answer a question and they popped off some value from *their* experiences. IMHO, It's not a…

Thanks for sharing! I will post this back to our content exchange too. :-) https://thwack.solarwinds.com/community/database-management/database-performance-analyzer/content?filterID=contentstatus%5Bpublished%5D~objecttype~objecttype%5Bdocument%5D

It has! You can download it from your customer portal or a trial from solarwinds.com.

Nice ideas augustinius. I've added a feature requests for you. http://thwack.solarwinds.com/ideas/4007 http://thwack.solarwinds.com/ideas/4008

nicolasdiogo, The link HolyGuacamole provided has the right info. Let us know if you need further assistance.

One theory... Since you're able to identify the blocking processes in SSMS, are all of the SPIDs under 50? If so, those are system spids which are filtered from some views. That might explain this.

SAM App Insight is more focused on system level performance metrics and DPA is more focused on query performance. The ability to answer why a query is slow and make recommendations are perhaps the most important differentiators. DPA breaks down query run time into specific wait events that indicates what was holding up…

P.S. You can read a story that illustrates how this would work here : Announcing DPA 9.2 GA : Is it the Application or the Database?. If you want to explore that for yourself, take a look at this AppInsight for IIS integrated with Database Performance Analyzer:…

+1 Jannis Griffin. Good to hear from you Dustin!

As far as what DPA can do now. You can create custom metrics and reports. So depending on what the SLAs were on, you may be able to achieve your desired outcome. I suggest having a call with an SE to discuss specifics.

Glad you figured it out. Been there... it's easy to forget such things when you don't do them every day. :-)

More or less. One means work is ready for the CPU but there are no OS threads to service the work. The other means that work is ready for the CPU but Vmware can't yet schedule the work onto CPUs.

Some more questions from one of my colleagues... Do you have any queries that explicitly call for recompile using "WITH RECOMPILE"? What version of SQL Server? Are you using an ORM like nHibernate? There's a great whitepaper from Greg Low on plan cache.…

We're unable to promise features to releases or dates. I can tell you we are actively working on it. It's not wishful thinking. :-)

I checked with support and they say you are correct. That dangling mi field isn't providing value here. It is possible to build an alert off of the tables that contain the high detail data but we haven't mapped it out. It may be complex and because the detail tables see more frequent activity, you'll have to carefully…

Well yes, they are there before the PLE drop, but what's important is what were they doing when PLE dropped. In order for PLE to drop, there must have been a request for a lot of pages not currently in memory. Unfortunately, we're not capturing any parameter values that may have been used in those SQLs so we can't really…

Have you tried using the Timeslice to go back to the moment that PLE plunged and see what was running? I've attached a screenshot showing how I can use Timeslice to get down to what was running in the couple of minutes where I know PLE dropped. There's 3 queries that had some significant wait time and I'd suspect them…