Data 'Ready-for-production' Checklist

If you've successfully connected your data source to Luzmo, it might be useful to go through this checklist to ensure that everything is ready for production! First we'll give some tips on securely connecting your data source, afterwards you can find some more general tips and finally some recommendations on ensuring performant datasets.

Security checklist

Do you use strong credentials to connect your data source?

Unique: Do not reuse passwords.
Random: A randomly generated password is ideal.
Long: The longer the password, the harder it is to brute force.

(Optional) Did you make sure that our static IP addresses are whitelisted in your data source?

(In case of a plugin) Are you validating the plugin’s app secret in all endpoints?

This Academy article explains how you can easily implement this validation in your plugin.

General checklist

Is all required data available in your Luzmo datasets?

If you want to visualize your revenue per month, you will need to have a column that indicates a date as well as a column indicating the revenue for that date.

Are all measures formatted to the desired format, precision and thousands separator?

Are percentages available as float numbers between 0 and 1?
Are dates stored as dates/datetimes/timestamps/...?
Are all numeric values coming through as expected?
Are all nested objects/arrays unpacked into columns/rows?

Is your data model ready to be used for displaying in dashboards and optimized for dashboarding by a non data engineer?

Have you mapped values and column names to business terminology?
Have you stripped non-relevant data (such as id’s used to link to data tables)?
Have you created logical, topic-based datasets to abstract the complex underlying data model from the dashboard designer?
This Academy course walks you through the typical analytical data modeling process!

Have you reviewed the dataset schema update options?

This article explains schema (metadata) updates in Luzmo.

(Optional) Have you linked the necessary datasets together?

This article walks you through the steps needed to link datasets in Luzmo!

Performance checklist

Is your database suitable for analytical queries?

Point 10 in the "Performance tips" article gives some recommendations on natively supported analytical databases.

Is your data model suitable for analytical queries?

Denormalize your data into a single view or star schema as recommended in Point 9 in the "Performance tips" article.

Did you connect a read-replica database to avoid resource contention?

Point 8 in the "Performance tips" article briefly explains why you would want to use a read-replica!

Is the read time on the dataset for your smallest and largest client within acceptable limits?

Have you avoided the use of derived columns by creating the needed columns from within the database source?

Point 5 in the "Performance tips" article explains when you should avoid using derived columns.

(Optional) Have you analyzed the need to materialize the data?

Point 4 in the "Performance tips" article explains when to use materialized views and how to set them up in PostgreSQL & MySQL/MariaDB. This functionality is also available in other data sources such as Snowflake, MongoDB, etc.

(In case of a plugin) Have you tuned indexes?

Point 3 in the "Performance tips" article shows you how you can analyze your query performance and tune the indexes.

(Optional) Have you enabled query caching?

Point 2 in the "Performance tips" article briefly explains how our cache works and how you can set it up. More information on caching can be found in this Academy article.

If you've successfully checked all of the recommendations above, your dataset is ready to be used in dashboards! If this is the first dashboard you'll be creating, the "Getting started" Academy course walks you through our platform and the dashboard editor. Once your first dashboard has been created, we recommend to go through the Dashboard 'Ready-for-production' Checklist to ensure it adheres to our best practices!

Data 'Ready-for-production' Checklist

Security checklist

General checklist

Performance checklist

Need more information?

Do you still have questions? Let us know how we can help.

Data 'Ready-for-production' Checklist

Security checklist

General checklist

Performance checklist

Need more information?

Do you still have questions? Let us know how we can help.

Related Articles