The past two years have seen the scale of Tracker implementations grow significantly (as was noted in our recent newsletter), with several countries now managing Tracker systems containing millions of unique individual-level records. In particular, countries using DHIS2 for COVID-19 vaccine registry or surveillance systems depend on Tracker to be able to enter, update and check the vaccine or test status of millions of individual people in real time, and to provide up-to-date reporting on these programs for decision makers and stakeholders.
Ongoing performance improvements in the Tracker application have helped make it possible for these systems to meet these demands, but specific implementation and configuration practices also play a very large role in enabling these systems to handle these large volumes of data. Performance optimization is thus a key component of supporting COVID-19 programs with Tracker.
Join members of the DHIS2 core team and HISP network for a webinar on implementation guidance for optimizing large-scale Tracker performance, with a particular focus on COVID-19 vaccination and surveillance systems. This webinar will take place on Friday, 21 January, from 14:00-15:30 Oslo time (GMT+1).
Complete this form to register for this free event, and you will receive a link to the webinar by email before the event takes place: online registration form
All registered participants will be invited to join the webinar on Zoom. You can also watch a livestream of the event and add comments and questions for the presenters here on the Community of Practice
Great, I can’t wait for this webinar. Our dhis 2 surveillance instance has become extremely slow. I hope we will learn some things that can help us improve the performance of our dhis 2 tracker.
@AL-Gassim Sharaf Addin The problem I faced with Tracker is often related to querying data concurrently from different program stages under event reports. I had to resort to API calls to achieve results. This route sometimes gets too hectic. I wished i could accomplish the task with just simple optimal steps.
Thanks for the feedback. I take it that you would like to be able to accomplish this in the event reports app itself instead of using the API? @dhis2-analytics is currently developing a new event reports app, would you be able to raise a Jira user story about this?
Would you be able to share more information about your performance issues with us before the webinar? Then we can make sure we have recommendations for your particular situation.
The performance issues we face relate much more to analyzes based on program indicators. We can no longer do certain analysis. for example the number of covid cases for all the period (2020-2022) which is based on a program indicator.
Analyzes are fast as long as they are based on data items only.
Here is some metrics :
Size of Guinea DHIS2 Surveillance db, with analytics tables = 41 GB;
Time it takes to run full analytics = 45min / 1h;
Number of records in datavalue table = 9 173 324;
Number of records in programinstance table = 2 253 382;
Number of records in programstageinstance table = 3 590 154
Number of records in trackedentityinstance = 2 274 700;
Bonjour chers tous,
Merci à l’équipe DHIS2 Centrale et le réseau HISP pour cette initiative. L’année 2022 commence bien avec le DHIS2.
M. Antoine ITOUA
In preparation for this webinar, the Tracker team has prepared written guidance on optimizing Tracker implementation at scale. You can read it on the DHIS2 Documentation site: Tracker performance at scale - DHIS2 Documentation
@maxk Thank you very much for this guide. It is really important. We have around 150 users logged in the last hour. I will be more interested in android app configs as we are constantly hit by database deadlocks while synchronizing. And we normally resolve this by restarting tomcat.
@marta and the android team will talk about android configurations on the webinar, but the deadlock issue needs to be investigated more. Which dhis2 version are you on? Do you have any monitoring tools the core dev team could get access to? If possible, please send us an email with details about this to performance@dhis2.org. DB logs, especially snippets that detail the deadlocks are very important.
Have you faced any particular challenges that you’d like to share? Thanks!