With the newSentryOne version 11 there's never been a more complete solution for our partners to provide the best service for their customers. In my prior role as a Strategic Partner I made use of SQL Sentry's tools to track what problems existed, and to point me towards their solution. With the ability now to monitor the entire infrastructure supporting Microsoft's data platform products, the ability to solve challenging problems dramatically improves.
Before I joined the company I knew that the capabilities of the tools exceeded all the other SQL Server monitoring tools. The concise dashboard with both the ability to examine performance metrics in real time as well as look at multiple years of history made it clear, seeing the SQL Server as well as the windows operating system metrics. By seeing this data in a single view, seeing memory usage at the OS level, and broken down in SQL Server by the different components within the SQL OS, by seeing wait stats, I/O counts, CPU percentage, batch counts, etc., all in a single pane of glass, I could more easily spot when a problem occurred, and could drill down to find its source.
What I learned since joining the company was how truly deep the tool allows me to investigate. I learned how I could use the Event Manager tool, not just to manage jobs, but to identify when problems occur, and easily see if one process (singularly, or repeated) is the source of the problem. I learned how to configure baselines, not just for a whole day, but for a variety of windows, so I can evaluate performance specifically during off hours processing based on normal characteristics for that window, or during normal business hours, evaluating those characteristics, or even at month-end, when the performance characteristics are quite different than a normal business day. Not only could I evaluate the performance based on those baselines, I could set up conditions to alert me when performance varies from the characteristics defined by each baseline.
In the last few years the database industry has moved away from relying on physical servers and into a world of virtualization. VMware and HyperV servers dominate in most company's data centers, and SQL Server instances primarily run on virtual servers using those virtualization hosts. The virtualization doesn't stop there, though, and companies are moving their server instances to the cloud, be it Azure or any number of other solutions. Microsoft also is providing cloud support for standalone databases in Azure SQL Database, and for their parallel data warehouse technology in Azure SQL Data Warehouse.
This year our tools expanded to incorporate monitoring across all these environments because, just like it made sense to see both the Windows and SQL Server counters in a single pane of glass, it makes sense to be able to see the metrics across the variety of environments supporting our business critical data. Without that visibility we have to guess where the problem really lies.
One additional feature we've added, in partnership with Tintri , is direct support of the performance metrics for virtual machines using Tintri devices on the back end, so you can see what the storage device reports for the I/O performance, and compare that with what the hypervisor reports, and what the Windows guest OS reports. This makes it far easier to know exactly where an I/O subsystem issue lies.
We also leveraged the data collected by our "old" Fragmentation Manager tool and built into the software the ability to evaluate what indexes are being used by a problematic query, what indexes could be used, what the ideal index for this query would be, and allow you to tailor and generate the most appropriate index for that query in your environment. We show you the statistics histogram in the same pane so you can see the data distribution yourself, to see what the most appropriate action should be based on the parameters supplied. This new feature has been built into Performance Advisor since the beginning of the year, but has also been released recently built into our free Plan Explorer product.
As partners, with all these new features, and with a reduced price from the previous Power Suite Plus collection, you have a complete solution for monitoring, diagnosing and troubleshooting your clients environments. You can configure the monitoring service at your client sites to send the data to a repository in their environment (if they wish to see the data as well as you), or you can configure it to send the data to a repository in your environment, and be able to view all your clients in a single dashboard view. That decision is best made between you and your clients, but that flexibility has always been there. In addition to the local repositories, you can also send the data to the SentryOne cloud, where you can view the system health and performance metrics via a web browser on any web-capable device.
What I found when using the tools at my client sites was that I could find and fix problems far more quickly and easily. This made my customers happier with me because it ultimately saved them money, and allowed them to use those funds for working with me to build better solutions to their business problems.
I'd love to hear your thoughts on the problems you've experienced, and how you've used our tools to solve those problems. Please, share those stories, and let me know if I may share them here. When I was a lone DBA years ago, I often had to define solutions to the problems that occurred. As my network grew I found I had resources outside my company who'd been through many of those problems. If you're willing to share your solutions, and how you did use or could have used our tools to solve them, we can all learn together from your experiences.
I look forward to hearing from you.
Allen