Introduction: Lotus Notes/domino developers who do not want to hear the following comment: "This application is great, but too slow!" In this two-part article series, we'll explain how to build a Notes/domino application that optimizes performance to avoid embarrassing situations like this.
One of the biggest drawbacks we know about a great app is that it's too slow to use--it's a waste of time and hard work for users because of their very slow response times. Over the past 12 years, we've spent a lot of time researching and testing Domino applications and their capabilities to understand how to optimize performance and which features are best suited to optimize performance. We started supporting and developing Domino applications early in the 1990s and were quickly attracted to the performance of the application. Many of the performance issues we were (still doing today) think of as servers are actually application performance issues. So the solution is often found within the application, not on the server.
In this two-part article series, we'll share with you some of the things we know. This article series covers three aspects of application performance: Database Properties, collection of documents, and views. In the 1th part, we'll explore database properties and collections of documents. In all cases, we will point out the most important places and provide concise examples of reality to help you understand how to optimize your applications. We'll use samples from a number of applications, and in these examples you can find at least one similar to what you're doing now or using. Our goal is to help you build both good and fast applications.
This article assumes that you are an experienced Notes/domino application developer.
Database Properties
There are database properties associated with application performance.
Do not maintain unread markup
If this check box is selected, unread markup is not tracked in the application, regardless of the settings for each view. We use Client_clock to track the time it takes to open a database, and our findings are astonishing. For a large application, such as an 200,000-document GB application, the Notes client can open the database (including network transports) in about five seconds when the unread markup is not used. If unread marks are turned on, you must wait six more seconds or more. These extra hours are spent on get_unread_note_table and Rcv_unread. If unread marks are turned off, these calls are not made.
In a smaller database (less than 1 GB), turning off the unread tag can save about 0.5 seconds of time. Of course, the time to open a small database is faster, regardless of whether the unread markup is turned on, compared to a large database. So before you put your application into operation, you should consider whether you need to use the Unread markup feature.