From the Blogosphere
Intuit QuickBase: Best Practices for Application Clusters
Effectively sharing data across applications
Aug. 7, 2014 03:24 PM
Data Sharing Features
Detailed below are three main ways that applications share data. The purpose of this article is not to detail the technical configuration of these features, but to explain their intended use, benefits, and risks.
Cross Application Relationships (Cross App)
Cross Apps allow a table in one QuickBase application to be shared to another application. The shared table, in effect, acts like a table in the shared-to application, even though it actually resides in another application. This has the benefit of allowing data into a single table and not having to re-enter or synchronize the data.
Table-to-Table Imports (TTI)
TTI are imports from one QuickBase table to another. TTIs can be within an application or across applications. The structure of the import can be saved and run manually. There is also an API, so the import can be programmatically executed. This is a great tool to synchronize data if a Cross App is not ideal.
Report Links are used on QuickBase forms to show data from other tables that are related by some key information. In most cases these are directly related tables, but they can also be configured to show data in other applications. This can be a good way to leverage organizational information from various applications on a rich form.
Application Clusters 101
QuickBase operates on a unique and proprietary technology framework. Applications run on a single server process that often has other applications running on it simultaneously. When data sharing features are used they cause the connected applications to act as a single application. This provides the benefit of having access to all of the data in memory and essentially functioning like a homogenous entity. The downside of this is that these application clusters can get very large and complex and may lead to performance challenges. They can also be quite difficult to unravel.
Over the years of developing QuickBase applications in highly complex environments, as well as dealing with some complex applications that have run into clustering issues, we feel a few common sense practices can help ensure the proper benefits of data sharing where appropriate and avoid risks.
- Make sure to plan and understand the use of data sharing features as well as clustering risk as part of an overall data architecture.
- Keep application clusters to well-defined and managed groups of applications that have a clear benefit from data sharing.
- Have other options available, such as programmatic synchronization of data using QuickBase API's where clustering may present a risk.
- Communicate clearly to application developers the risks associated with data sharing features and have a process to review and approve their use.
- Periodically check applications for clustering by adding ?a=listdependentdatabases to the url on any application home dashboard.
- Be cautious when copying applications as Cross Apps may be retained and Report Links will be retained which could cause a substantial increase in cluster size.
- Be very careful when adding previously built applications to a cluster as they may have complex networks of sharing in place that could suddenly grow the cluster far beyond what is expected.
- Understand that TTI will create what is called an Intermittent Dependency so the clustering only occurs when the TTI is run and will eventually decluster. This can help avoid clustering if planned properly during off-use hours, but can also be a source of unexplained performance issues if not managed properly.
MCFTech is a leading QuickBase Partner and highly experienced at handling complex scalability and system design challenges. For more information please visit our website at www.mcftech.com or complete our contact form and a client solutions representative will reach out to you.