Transcript

SQL Diagnostic Manager is a powerful performance monitoring learning and diagnostics solution that proactively notify administrators of health performance or availability problems within their SQL Server environment via a central console web-based dashboard or mobile device. It minimizes server downtime by providing agentless real-time monitoring and customizable alerting for fast diagnosis and remediation cycle. SQL Diagnostic Manager support all SQL Server versions from 2000 and above. The collection is centralized in […]. Nothing is installed on your server. You can adjust your collection cycle as slow as one minute. All data collected is stored in a customizable historical repository that allows you to view from the console what occurred in the past or run historical reports. We provide you with three main consoles to view the collected data. The first one is a thick console, a .NET console installed on your computer or laptop that allows you to configure, manage, and view real-time and historical performance data and create reports. The second one is the web console. It is a web browser-based console allowing you to see the performance of your instances. And the third one is a single DM [SQL Diagnostic Manager] mobile, a mobile-friendly web interface for your smartphone or tablet devices. You can organize instances in views, a logical grouping of instances. By default, you will have views for all servers’ critical warning informational. In any server in maintenance mode, this allows you to focus on a particular sector of your environment or instances with a certain status. The quick at a glance box gives you quick information on your instances. Hover your mouse to see top alerts. See quick information on the different areas of resource usage for your SQL instance. Tags are attributes you apply to your instances. You can assign multiple attributes to one instance making it easier to find instances or do certain activities in bulk. For example, if I want to upgrade my virtual environment over the weekend I can right-click the tack called virtual and set it to maintenance mode at a specific time. SQL Diagnostic Manager is also divided into the following sections: overview, sessions, queries, resources, databases, services, and logs. In Analyze you can see alerts from many areas within SQL Diagnostic Manager. Quick at a glance where you can hover your mouse can see detail information. SQL DM [SQL Diagnostic Manager] today where you can see the active alerts that you have for the day. And the alert history where you can filter by metric date severity and find any alerts that occurred in the past. Let us take a look at an active alert on their SQL DM [SQL Diagnostic Manager] today. I see that I have a blocking condition going on right now. Clicking on that alert takes me to the sessions area where I can see the blocking condition. From here I can get additional information from the blocking report. The blocking report is a snapshot in time of the blocking condition with additional detail information like the host, the user, the application, the database, and the last command that was issued. Showing blocking conditions in real-time is great but what if that condition already occurred in the past with SQL Diagnostic Manager. I can use the history browser to navigate back to when a specific condition occurred. Here I have selected a blocking condition for a specific server over the weekend when I was not in the office. I can right-click and show the historical view. And it takes me to that specific time and date. Here I can go forward or backward for each collection period. In my case is every minute. Deadlocks are also shown in a report format. Although deadlocks resolve itself in SQL Server we can still get a report on which session got terminated. We can export that information into an Excell file that can be easily shared with someone else. So they can open any Management Studio. In SQL Diagnostic Manager we can see a list of the top worst performing queries in SQL Server. First, you want to define what is considered a bad performing query. In my case, I am doing it by duration. And anything that is above five seconds. Then you can easily filter the queries you want to see either by time, application, user, or any other filter information. Here I can sort any column for easy identification of the problematic query. There are two main views: signature mode and statement mode. The signature mode shows all active queries group by the raw signature of this statement. You can see here that this type of statement when X number of times this helps me determine how often is this query hitting MySQL Server and if it is causing any degradation with the resources on my second server. The statement mode allows you to see each individual statement exactly as they were collected by the query monitoring trace query. The stats are shown in two different ways: over time and by duration. Over time you can see the time of the day associated with the weights and their total weight. On my duration shows you the biggest offender by total weight. I am allowing you to drill down to the statement, the application, the database, the client, the sessions, and the users. I can start my drill down from applications to see which weights are being generated from a specific application. In SQL Diagnostic Manager alerts can be enabled or disabled as you wish. Thresholds are set by default but can be easily adjusted. Another important concept is the use of baselines. Baseline defines their normal use of that metric for a specific period of time. You can customize baselines to define your peak periods or business hours. Baselines will show on this chart to guide you with the optimization of your thresholds. Or you can simply use the recommendations that we give you based on your normal use. Baselines are also used to alert based on a percentage above or below the current baseline. This will make that alert more dynamic as the baseline will fluctuate over time. There are several ways you can execute and respond to an alert. An email is one this will allow you to pass variables to your own customized template. And based on certain conditions like days of the week tag specific alert status you can generate an email. You can also run an executable or batch file. Or if you wish you can run an agent job by loading the job and even specifying a specific step within that job. You could also run a SQL script to automate a corrective action. In this example, if my transaction log gets full I can automatically run the script passing several variables and truncating my transaction block. There is also a tempdb information to see if there is any contention within the tempdb and the associated sessions using the tempdb on their table and indexes. You can see the size dependencies index. And in the statistics of a table. You can also see the fragmentation of that table and even issue a rebuild or object statistics right out of the console. In the analyze area foundations are given based on your selection of areas you would like to analyze in your SQL Server areas such as blocking process, deadlocks, resources, database, and server configuration index in query optimization to name a few. You are presented with a finding list of all the expert recommendations that we have found on the server. You can select each finding to see more detail information on why we issued a recommendation. Details about this recommendation in some of them. They include links to external articles to learn more about our condition. And also in some cases, you have a SQL script that you can run directly from the console to optimize the condition. There are over 30 reports that come out of the box with SQL Diagnostic Manager. You can run these reports on the man scheduled them to be sent on an email deployed into SSRS [SQL Server Reporting Services]. Or you can actually export them to Acrobat or Excel. For more information please visit idera.com where you can download a full trial of SQL Diagnostic Manager for 14 days and 15 instances. Thank you.