Data Pipeline

Collection #

We provide an API interface in order to collect data. Using our SDK you can strategically place API calls within your game where analysis is required.

public class Manager : MonoBehaviour
{
    public void PostAnalyticsEvent()
    {
        // Assumes MKNet Analytics has been initialized, somewhere in your start up sequence
        // Like this: MKNetAnalytics.Init();

        // These two calls are equivalent

        MKNetAnalytics.ReportEvent("important_event", new ImportantEvent()
        {
            Foo = "Bar"
        });

        MKNetAnalytics.ReportEvent("important_event", new Dictionary<string, object>()
        {
            {"Foo", "Bar"}
        });
    }

    private class ImportantEvent : ClientModels.Analytics.AnalyticsEventData
    {
        public string Foo { get; set; }
    }
}

Processing #

Our servers sort and validate event driven data into clusters and tables and distributes them to dedicated PostgreSQL servers.

Storage #

We have developed a data warehousing model that takes advantage of very powerful PostgreSQL features like binary JSON, Materialized Views and Clustered Indexes.

Normalization #

As a final step we deconstruct the data into queryable views consumed by third party visualization tools like Tableau.

MKNet

Related API Endpoints #

Loading API spec...