Transcript

Expand

In this session, we will talk about Precise for SQL Server. It is a member of the Precise application performance platform group of products offered by IDERA. The key benefit of Precise is that it offers high-resolution visibility into SQL statement execution by sampling at a minimum once per second. And we correlate cause and effect. So you always know exactly what happened to your system. Precise tracks all execution states including locking wait states including use of storage. Precise offers alerting, reporting, and dashboards to communicate with other stakeholders within the organization. Precise will automatically identify performance bottlenecks and offer recommendations. There is a long term history the offers trending. Precise captures the execution plan which connects SQL statements to the tables and indexes. Therefore we can turn our data upside down. And we say well what are the busiest objects on your system. And that is going to be crucial for knowing about your defragmentation. Routines will also offer index recommendations and modeling to minimize the risk of a production change. And also to make sure that you have the right set of indexes for your system. Will recognize when indexes are unneeded. Precise tracks at the operating system including a plugin for VMware. And so Precise can be used for capacity planning. But with Precise, it is not always about more CPU [central processing unit] and more memory. It is also about processing efficiency. So let us let us go right in. This is the Precise initial screen. The dashboard here we would show every instance by availability execution count. And you can see that over the past two days we have had 652,403 executions. This is a bar graph segmenting where the time was spent. Now over here we have more details about this particular instance. And the y-axis represents in MS SQL time. And as you can see it is segmented into these execution sub-states. This is 100% comprehensive. The x-axis is time across the day. It is very easy to just select a time frame. Or manually update the time frame to include the processing window that you care about. And when you do that Precise will adjust its findings report to say well what are the biggest resource consumers on the system. And we are going to recognize heavy objects as well as SQL statements including environmental things. So here we have high CPU [central processing unit] weight. And I could immediately hot link directly to CPU [central processing unit] pressure. But before I do that let us click learn more. Precise is going to give me information about what to do next. And it is going to show me how to alleviate CPU [central processing unit] pressure on my system. I want to draw your attention over here. Because this is our complete library of findings and recommendations. And it addresses just about every circumstance that you could encounter within SQL Server. So this can be very helpful for elevating newbie DBAs [database administrators]. As well as giving a new idea to seasoned DBAs [database administrators]. Close that out. And go directly into a long-term report. Here we are showing processing day by day during March and April. It is very easy to spot that (you know) we have a burst. A burst system. It is very easy to spot particular days when processing is way above the baseline. That is that light blue line. There is a dotted line that represents the exception threshold consulting the actual value versus the baseline over the past month. I can see that we are significantly above the baseline. Now please make a mental note of this profile. Because we are about to drill in and highlight links directly from the report to the expert view. This is a distinct advantage of Precise is the alert reports. Expert view are all correlated together. And that includes dashboards. Because there is one central historic repository that we are using to show the data. Clicking this hot link. I now go in. And first of all I am able to see additional details. Like on April 25th I can see that there was locking. Down here I can see well what databases drove this profile. I can select the TPCC database. And now I have isolated just the SQL statements executed by that database. And they are sorted in descending order by an MS SQL time. This takes into account the number of executions. This is where Precise stands out. For low latency high throughput applications that is (you know) kind of (you know) your OLTP [online transaction processing] applications. You are very busy with finance. sa [system administrator]. PeopleSoft applications precisely is going to give you the true profile. And it is going to show you where those SQL statements are spending the time. Dark-blue consulting the key that shows us. It is used in CPU [central processing unit]. We do not want to just focus on locks and waits. Because that would actually deny us information about what is happened with particular SQL statements. And where there is a potential buyback by minimizing execution time. I minimize the opportunity for contention blocking its etc. And so this path to performance helps the DBA [database administrator] take the instance, the database to the next level. Now you can see that there is a number of tabs along the top. Here want to take a look at objects. This is another distinguishing factor of Precise. We are going to track the database. And track changes. We are also going to track the use of space down here. And mention this earlier. This is where we show active objects. The water line table is probably 10 times busier than any other table in the system. I want to make sure that this table is in high performance by making sure that it is included in my maintenance routines to reorg rebuild minimize weight space. So I make every I/O [input and output] of the blocks and leaf pages as efficient as possible. Now additionally I can investigate this order line table. And a couple of things here is. We are going to track its number of rows, its space. But down here I can correlate to the SQL statements that use this order line table. This is critical in business applications. You are going to start to see a commonality in the where clause order by the sort by. Because think about it in a business environment. People want to know and report one well what happened last our last day, last week, last month, last quarter, last year. And so there it is going to naturally be a commonality in these SQL statements. Now what you can do is. You can use Precise to recommend index or different better indexes. And here is showing to create four indexes. Precise will actually recommend index drop it dropping indexes as well the key is this. This evaluate button applies these changes in this what-if recommendation to basically say well what would happen if these indexes were applied. And this is our modeling capability. And what it shows is that these SQL statements would be impacted this table. SQL statement had consumed one hour four minutes that is 64 minutes. And I save 98%. The next SQL statement down consumes 27 minutes. I save 1/3. That is nine minutes. So I have already saved over an hour’s with the processing. And if you can consider that each core runs at millions of instruction per second. This is going to have a nice ripple effect across the SQL state SQL Server database instance VM. As well as if you have other VMs running on that server. It is going to impact them as well. Because you are saving resources by improving processing efficiency. And this is where Precise really head and toes above any other solution on the marketplace. So just as a quick recap. Always on 24 by 7. This is not one of those tools where you have to turn on additional visibility. We provide all of the information all the time. As well as alerting to automate monitoring of your system. Let me know when a problem is developing. We also track the physical server, guest VM, the host. And that gets into capacity and processing efficiency. So you have both sides of the coin. It is comprehensive. Precise allows me to drill down. You saw by the database. But I can also go in by login, program, by machine. Precise offers recommendations would with what-if modeling. This minimizes the risk that you are going to make a negative change in a production system. Precise offers dashboards and reporting for operational awareness to share information for all stakeholders. If this has intrigued you please visit IDERA’s website. Go to the products page. And these are under the Precise solution. So clicking on Precise for Oracle will bring you to the Precise application performance platform group of products. And click See In Action. And that will bring you to a sign-on screen. And we will talk soon. Thank you very much for your interest in Precise for SQL Server. Thank you.

Topics : Database Diagnostics,Database Monitoring,Database Performance,

Products : Precise for Databases,

An Overview of Precise for SQL Server

Precise for SQL Server is designed specifically to keep your SQL Server database environment available and performing at peak efficiency. It captures, measures, and analyzes performance metrics from all critical system components and helps you detect and correct the root causes of problems before end users are affected.

Precise for SQL Server provides high-level visibility across multiple SQL Server instances and provides alerting to problems, reports on collected data, and maintains a warehouse of SQL Server database performance information.

Precise for SQL Server continuously analyzes database behavior to find the poorest performing objects or processes. Historical statistics stored by instance, user and program provide the necessary context to help further DBAs in seeing the entire picture.

Precise for SQL Server quickly analyzes SQL statements, database objects, and SQL Server instances to uncover specific causes of performance problems. It presents the SQL Server access path in detailed steps and displays the statistics needed to understand each step to isolate problems fast.

Once a performance problem is identified, Precise for SQL Server allows DBAs to drill down in the context of the user transaction to pinpoint the root cause of the problem. Learn More →

facebook  
Contact IDERA: