dbWatch Powertools: Database Monitoring

The secret to monitoring data is about finding the perfect balance. Too much data and you’re drowning, not enough and an ailing instance may slip under your radar. The perfect mix will track your instance’s vital signs, putting your finger on the databases’ pulse. With these vital signs, you can quickly find out if the database is healthy, while being able to track and understand any irregularities.

The secret is to find the sweet spot with the perfect amount of data. At dbWatch we have spent years finding it. Everyone can monitor 10 – 50 instances but if you have 500 there are very few applications that handle those numbers efficiently. At dbWatch, we are constantly developing and perfecting our product making a very effective, cost efficient solution. Our customers say that we are one of the best on the enterprise level of monitoring.

What dbWatch does / Tracking and Monitoring

dbWatch software tracks information so you know what was done, and when it happened. This data helps you decide how to allocate future resources, for example – knowing how capacity will increase or decrease and making changes accordingly.

Monitoring allows you to see session history and to better understand its patterns. dbWatch monitors the important information so you can constantly know that status and health of your database. We try to give you enough information without overwhelming you with it. Alerts are used for more pressing matters which are very important for you to investigate.

Matching Databases with Applications

Data needs to be viewed efficiently. By looking at the resources an instance consumes, you can see if there are enough resources available for the instance, based on the hardware components. When we see the big picture of how an instance uses time, memory and resources, we can easily know if the instance is working efficiently.

You can also use that information to match the database with the applications. To give an analogy : if databases were cars, applications would be drivers. Consider the Mini Cooper. It’s a smart little car, but even with the best racing car driver, it won’t win a race. On the flip side, a Ferrari driven by this blog writer would not win any races either.

To have a system running optimally, we must make sure that the car and driver match. Monitoring information will help define what is needed and ensure that you get a Formula 1 Mercedes driven by Michael Schumacher when you need to race. However, if you just need milk at the store around the block, a mini cooper, driven with a careful driver, would be more logical.

User pattern

The number of users on the database also gives insight. dbWatch tracks long term user patterns. These give insight into the database. Returning to the car, if something goes wrong you can see the activity and resource consumption and when the problem happened. Did it happen at high speed / peak database usage? Or perhaps it happened when the car was parked.

This also is useful when deciding about making changes to your databases. Perhaps what was once the ‘milk run’ to the store, now has many users and needs a ‘bus’ instead of a ‘car ’ because the database is much more active. The user patterns help you see these changes and make informed decisions about what, if any, action to take.

When you can see what isn’t in use, space can be cleared for higher priority active instances. Importantly when an instance’s use in minimum, it can be taken offline or made into a read only document.

Growth rate

Every finance department likes a good budget and a careful prediction of growth rate can make you friends in finance. With a few years of tracking disc usage, you will have a very thorough snap shot of size. For example, when usage increases five percent each year, you know what to expect from that instance in the next year. This plan allows you to know how much space is needed for the coming year. You can then project the budget and make your counterparts in finance, happy.

Maintenance routines monitored

We are all guilty of deploying and forgetting – making a maintenance routine and then leaving it to work on its own, without checking in to see how effective our routine is. However, it’s important to make sure they are working after they have gone out into the ether. Some need almost no attention, while others require adjusting and tuning before they are optimal.

These are some of the key points that dbWatch can do for your databases. Email us to get further information and a free trial of our systems.