System Memory

The System Memory section displays information about the amount of memory being used by different processes on the server as well as page faults and page file usage.

Disk IO

The Disk IO Section displays the read and write latency for each of the physical disks on the server.

Metrics for all SQL Server or SSAS instances on the server are integrated into the Windows Network, CPU Usage, and System Memory charts, allowing you to quickly determine exactly how much of these resources is being used by each instance, at any point in time.

SQL Server Activity

The SQL Server Activity Section displays information about what the SQL Server instance is doing.

SQL Server Waits

The SQL Server Waits section displays information about the classes and categories of waits that occurred as well as the duration of milliseconds that the waits were in effect during that time period.

SQL Server Memory

The SQL Server Memory section displays information about how the Server instance is using memory that has been allocated to it. 

Database IO

The Database IO Section displays information about the read and write latency for the databases.

Understanding the Dashboard

For more information about the various performance metrics displayed on the Dashboard, see the Performance Metrics topic.

For performance reasons, the Dashboard never queries the remote server directly; all data is retrieved from the SQL Sentry database after it's collected by the SQL Sentry monitoring service. 

When in real-time mode (auto-refresh is enabled), the charts and gauges update automatically approximately every 10 seconds to show the values for the last collected sample. Enable real-time mode by selecting the Play button in the toolbar.

Dashboard Display Modes

The Dashboard has two display modes: Sample and History. The display mode can be changed at any time by toggling the mode toolbar button.

Sample mode uses various bar charts and digital gauges to show activity for a specific time, or sample time. This mode includes certain metrics that aren't available in History mode, such as Network activity by adapter, CPU Usage by individual processor, and SQL Server Waits by specific wait class.

Note:  In Sample mode, the CPU Usage chart displays Kernel time in red, and Processor time in green for each of the cores on your monitored instance.

History mode uses line and stacked area charts to show you how the Windows Server and SQL Server targets are performing over any time range. The range can be changed by adjusting the Start and/or End times on the toolbar, and then selecting Go.

Customizing SQL Server Waits Chart

One of the most common reasons our customers customize the SQL Server Waits Chart is to break down the Other category into something more specific. 

Customize the SQL Server Waits historical chart by adding an additional category to the Performance Analysis Wait Type category table, and then assigning counters in the Performance Analysis Wait Type table to the newly created category.

For example, create a new category for parallelism by using the following query:

New Category Creation

INSERT INTO [SentryOne].[dbo].[PerformanceAnalysisWaitTypeCategory] (Name, SortOrder, Enabled, RGBColor) 
			VALUES ('Parallelism',6,1,'#bb7fd1')

After creating the Parallelism category, add the following counter to it:

Adding a Counter to the Category

UPDATE [SentryOne].[dbo].[PerformanceAnalysisWaitType] SET WaitTypeCategoryID = 7
			WHERE Name = 'CXPACKET'

Interacting with the Dashboard

Changing the Active SQL Server/SSAS Instance

You can change the SQL Server or SSAS instance using the Instance drop-down control in the top middle of the screen. The metrics in the right pane reload, but the active date range and other options are maintained.

The following functions are available in History mode:

Tooltips

Hover over any point on the chart to see tooltips with specifics about the sample, including the sample instance Time, Name, and Value.

Select any graph metric to highlight that metric on the graph.

Sample Selection

Clicking on a history chart presents a vertical red line at that position called a cursor. The selected time is indicated by the Sample Pos box at the top right. The cursor is synchronized across all charts allowing you to easily correlate what's happening with one performance metric with what's happening with other related metrics at the same time.

After selecting a sample with the cursor and toggling Mode , detail data for the selected time loads in Sample mode, allowing you to see specifics for that sample that aren't available in History mode. For example, if you select a spike on the SQL Server Waits chart and change to Sample mode, you can see the specific wait classes responsible for the spike, whereas in History mode only general wait categories display.

Click and Hold

If you click and hold a sample, the cursor changes color, and when you release the mouse you'll be presented with a subset of the Jump To context items used for range selection. This allows you to jump directly to a different tab and view only operations that overlapped the currently selected sample time.

Range Selection

Select a history chart and drag the cursor to select a range. Range Start and Range End boxes update, and releasing the mouse button causes context items to appear, allowing you to choose whether to Zoom In the selected range or Jump To another part of the application using the selected range.  For example, selecting a range that includes a spike in network activity, and then selecting Jump To > Top SQL, displays and loads the Top SQL tab with queries that ran during that range, enabling you to determine which queries may have been responsible for the network spike.

Advisory Condition Markers

Display Advisory conditions on the Performance Analysis Dashboard graphs by enabling Advisory Condition Markers.

Select any Advisory Condition Marker to see the Advisory condition(s) that occurred.

Colored lines across the bottom of each graph indicate where a condition fired, and the duration a condition was in effect for each instance. Hover over a colored line to display this information.

 

Selecting the name of the condition takes you to that event in the Events Log.

Toolbars

Sample Mode Toolbar

Sample mode button	Description
Show History Mode	Changes the view to History mode.
Quick Trace	Runs a Quick Trace.
Show Windows Processes	Displays Windows processes running on the server.
Jump To Calendar	Opens the corresponding Event Calendar.
Report	Opens the Report viewer for Windows and SQL Server performance data.
Show Advisory Condition Markers	Displays Advisory Condition Markers on the Performance Analysis Dashboard.

History Mode Toolbar

History mode button	Description
Show Sample Mode	Changes the view to Sample mode.
Run Quick Trace	Runs a Quick Trace.
Show Windows Processes	Displays Windows processes running on the server.
Jump To Calendar	Opens the corresponding Event Calendar.
Report	Opens the Report viewer for Windows and SQL Server performance data.
Show Advisory Condition Markers	Displays Advisory Condition Markers on the Performance Analysis Dashboard.

Controls

Sample Mode

Network    

• Total - Displays network activity as an aggregate across all adapters, with SQL Server/SSAS activity integrated.
• Adapters - Displays activity by the individual network adapter, enabling you to spot adapter-specific saturation issues. 

History Mode

Disk IO

• Total - Displays average latency across all physical disks.
• By Disk - Displays actual latency for a specific physical disk.

Database IO

• Total - Displays average latency across all database files.

• By Database file - Displays actual latency for a specific database file.

Using Quick Trace™

Quick Trace gives you an instant snapshot of what all of the processes on a SQL Server are currently doing. Trace and process-level data is collected for a specified interval and then automatically correlated, effectively combining the functionality of Profiler and Activity Monitor together with a single click. Launch Quick Trace by doing any one of the following:

• Select the Run Quick Trace toolbar button.
  
• Select the Quick Traces tab, and then select Run Quick Trace.
• Right-click any chart in the Dashboard, and then select Quick Trace.

For example, if you observe a spike in activity while in real-time mode, you can immediately launch Quick Trace and see exactly which SQL Server processes or applications are responsible for the spike.

Viewing Windows Process Activity

See metrics for all Windows processes using Show Windows Processes directly from SQL Sentry, which avoids having to open a remote desktop connection to a server, and open Windows Task Manager. The Process Activity window loads all active operating system processes, sorted in descending order by CPU by default.

Launch the Process Activity window by doing any one of the following:

• Select the Show Windows Processes toolbar button.
  
• Right-click a target in the Navigator pane, and then select Show Windows Processes.
  
• Right-click any chart in the Dashboard, and select Show Windows Processes.

Event Overlays

Specific events taking place in your environment can also be overlaid on the Performance Analysis Dashboard. Access this feature by selecting Jump To Dashboard from a number of places within the SQL Sentry client that display a range of historical events, including any: 

• Event Calendar
• Top SQL
• Blocking
• QuickTraces

When using the Jump To Dashboard feature, events are overlaid on the x-axis for each chart. The start of each distinct event is represented with a green circle, with the event's duration represented with a pink line, ending in a pink dot. Hovering over an event displays a tooltip about that event.

From Event Calendar

When using the Jump To Dashboard feature from Event Calendar any events that are highlighted with the Calendar Highlighting feature are overlaid on the Dashboard.

From Top SQL 

When using the Jump To Dashboard feature from Top SQL , if the Show Totals is enabled all events belonging to the selected group are overlaid on the Dashboard. If Default mode is enabled only the selected event is overlaid on the Dashboard.

From Blocking or QuickTraces 

When using the Jump To Dashboard feature from Blocking or QuickTraces the selected event is overlaid on the Dashboard.

Other Performance Analysis topics

For more information about general Performance Analysis features, see the following topics.

• Baselining

• Dashboard retention resolution

• Performance metrics

• Performance Overview

See Performance Analysis features available for your monitored target type

For more information about Performance Analysis features for specific target types, see the following topics.

• SQL Server

• SSAS

• VMware

• Windows & Hyper-V

• Azure SQL Database

• Azure Synapse SQL pools

• Analytics Platform System