Query Metrics to Help Optimize Database Applications

There are many different factors that affect the performance of an application. For instance, newer programs tend to need newer and faster computers to run properly. A computer with an older graphics card would run newer games very slowly, or not at all. Another factor would be the efficiency of the application itself, its design, and coding. In this case, the use of query metrics can help to streamline an application to make it more efficient and perform faster.

Ideally, the efficiency of an application is optimized at the design stage, when the application is just being created. It is much cheaper and easier to make changes to an application at the conceptualization and coding stage. However, in most cases, an application that needs to be made more efficient is already in existence. Furthermore, usually the application is already under use, and a total redesign is not an available option. In this case, it becomes important to be able to identify which areas or aspects of the application are contributing to its inefficiency.

Now, for database applications, no matter how complicated its structure might be, the core performance issues usually arise at the database level. That is, the efficiency of database access more or less determines the overall application performance. Accessing the database, especially if the database is very large and accessed many times, is the most demanding task.

It then becomes important to consider the relevant metrics regarding queries in order to trace database performance. By looking at how well the various running queries are performing, it becomes possible to determine whether any inefficient behavior is occurring. It also becomes possible to pinpoint any inefficient behavior right to their source, to the type of queries that slow down database performance.

One common problem that arises in SQL databases is what is known as "session blocking". This occurs when two separate SQL sessions attempt to access the same block of data simultaneously, while also putting a lock on this data. Then one session would have to wait until the other releases the data to be accessible to others again. Session blocking can be determined by querying the sysprocesses table. Patterns and the various metrics can be obtained, and the session blocking behavior can be traced to its root.

When a particular inefficient query has been discovered, its cost and the CPU time and disk input/output used to execute it should be considered. Query cost is usually available from within the program used to run it, and calculates the performance cost of a query based on its component steps. The same is true for the metrics CPU time and disk access: these statistics are often available using the same command or option in the database program.

These query metrics are essential to consider to speed up inefficient queries. Efficient queries would lead to efficient database access, and hence to fast and efficient database applications, such as servers. Optimizing queries is relatively cheap, requiring no new hardware or much capital outlay, and can be done even with applications already being used

Query Metrics to Help Optimize Database Applications

How to Create and Maintain Database

Your database is like your house. If it comes crumbling down, you have no home. If it is solid and built properly it will stand the test of time.

Your database is the single most important asset you own as a business person. Without it you have nothing; with it you have everything. Even if your building, items, computers, etc were to burn down around you, if you still have your database, you still have a business.

Creating your own database.

I am amazed at how many business women do not have a database or at least a structured set up to maintain the list that they have.

Imagine if you have 500 contacts all strewn around the house on business cards, address books, in shoe boxes or in digital format. How can you manage and market with that?

And what about if you have only a few contacts and are wondering what point is there? Well, there's plenty - Imagine the work you will have to do once your contact list reaches 500. It's much harder to manage it then if you were to begin now.

Begin with these -


Write a list of all the people you know
Add your relatives, friends, work colleagues, organisations, church etc
Put them in an excel spreadsheet neatly classified so they can be sorted
Or - enter them into outlook or a professional database like ACT
Or list them in your web based data base

You now have your first list.

What do I do with the database now?

Now you can begin to market to your list.

Write up a plain text email letting everyone know you are in business, add a special offer, ask them to send it out to their friends and voila - you are now e-marketing!

What if I want it to do more?

Once you find out what is possible I guarantee you will want to do more, this is why it's best to start with a professional database.

With these you can sort and divide your list into different categories. i.e. you might only want to market to people who are not a current client. Or you want to send a reminder to the people who are attending a workshop you are running.

Having the ability to sort your database gives you the opportunity to begin e-marketing with a targeted approach.

How do I make sure I don't lose it?

This is a great question - it's vital that you back up your database every week. I have a reminder on my calendar each Monday to not only back up my database but all of my important files on my computer. You can use an external hard drive, but make sure you keep it away from your computer. I once had a friend who kept her external hard drive that she backed up on in her laptop bag. The bag was stolen along with her backed up drive and her lap top; her entire business gone.

If you have a customer management system tied in with your website, the database is stored on the server so it will always be retrievable.

How to Create and Maintain Database