“The system is slow” can be an irritating and often unfounded user complaint about Alfresco or Documentum. To address user concerns, Alfresco or Documentum support resources have to often become detectives and ask questions like “what were you doing, which documents, what time, from what machine….” to try and diagnose the background of the complaint. Users can get frustrated as typically the compliant represents an overall feeling rather than any exact action or time. This post will present how the latest release of OpenAnnotate looks to address the issue of user performance complaints by giving the user a clear understanding of the components of document viewing performance and actual timings to pass along to support resources.
Alfresco or Documentum – Viewing Performance Components
Viewing documents in Alfresco or Documentum with a browser based application will include the following performance components:
- Time to login to the system via the network
- Time to load the javascript in the browser for viewing the document
- Time to retrieve the document from Alfresco or Documentum
- Time to load the document into the browser via the network
- Time to display the document
For Alfresco or Documentum support resources, typical performance inconsistencies can be tied to
- Network performance – Users accessing the system from overseas, cellular connection or via a slow VPN will see dramatic performance issues. This is an area that is hard to address by support resources.
- Document Size – Large files take longer to transmit and view than smaller files by both the network and the browser.
- Browser/PC/Device Performance – Viewing a document within the browser requires the user’s device to present the document. A faster device will display the document faster.
Alfresco or Documentum performance – Whether tuning is required or additional capacity needs to be added on the server (perhaps during large document batch loads) can lead to performance issues that system support resources should be addressing. See related post on Alfresco Health Check
To address user performance perceptions, the application should be able to educate the user and identify network and device performance rather than blame all performance issues on the Alfresco or Documentum system. With a proper understanding of the components of performance, complaints can shift from “The system is slow” to “I need a new computer – it is slow” or “The network is slow” resulting in better perception of the Alfresco or Documentum system as well as less Alfresco and Documentum support complaints.
OpenAnnotate – Updating User Perceptions on Performance
The newest release of OpenAnnotate includes a “Load Time” performance timing accurate to the hundredth of a second.
The load time is calculated from when OpenAnnotate is first called (View or Annotate a Document) and the image of the first page of the document is displayed. If a user places the mouse over the load time, the complete performance components of the load time are displayed that are color coded (Green/Yellow/Red).
The option of copying the performance time to the clip board to share via email/slack or other method is available for the user to report performance concerns.
When the network is slow, users will be able to clearly see how network performance has affected load time. Here javascript load time (heavily influenced by network speed) takes notably more time.
Displaying the load time addresses user performance concerns in that:
- More Detail – Instead of just stating that “the document took too long to display” the user can be lead to say that the document took “x.xx seconds” to display to provide more context on performance.
- Network and Device Understanding – From the detailed review, users can clearly understand the network performance and device performance versus the system performance. When concerned about performance of any component, detail can be sent to the appropriate support resource.
See it in action here :
Overall Alfresco and Documentum Performance
Both OpenAnnotate and the OpenContent Management Suite log performance numbers leveraging the Big Data ELK Stack (see related blog post). The ELK Stack (Elastic Search, Logstach and Kibana) provide a way for support resources to monitor all user performance to spot performance trends. By aggregating the performance numbers, support personal can tell
- Are there certain users that have worse performance than other users? Analysis could lead to device or network upgrades.
- Are there certain times that performance is worse? Analysis could lead to changes in Alfresco or Documentum or network issues.
- Are there certain types of documents that cause performance to be worse? Analysis could lead to tuning for larger documents with object store or network performance.
Look for a more detail post later in August.
Summary
Changing Users’ perception of performance requires educating the user to what components can affect document viewing. OpenAnnotates’ newest release includes load time performance and timing components to help educate the user in to avoid the “system is slow” complaint.
Let us know your thoughts below:
[…] Performance metrics for OpenAnnotate […]