FIG. 20 illustrates an exemplary process of collecting and sorting data for sensors such that display over time could be fast and efficient in accordance with one or more embodiments. The best way to depict this approach is by using an example of one sensor reading where the update rate is set to every 30 seconds. This would lead to 24*60*2=2880 data points per day, and that would lead to average of 30*2880=86400 data points per month and 12*86400=1,036,600, more than a million data points per year. If the request from the user is to display multiple sensor readings over a year, getting a million data points for each sensor would be very inefficient and would cause delays in retrieving and displaying such data. This would result in user experience that is not desirable. In order to improve this experience, a filter or buffer is designed in the cloud architecture such that data is pre-processed for the most optimal experience. The data is averaged on hourly sensor readings and stored in a separate table, averaged daily for all hour readings, average weekly on all daily readings, and finally average monthly on all daily readings. Having multiple tables would allow the end application to retrieve the data more efficiently (e.g., only 12 data points for the entire year or 365 data points for the entire year). This would result in an improved user experience and reduction in data display latency. This down sampling of data approach further improves user experience.