As we approach the end of the year, we are taking a look back at some of the perennially useful developer articles from years past. They might be oldies but they are still gold. This article was first published on the Cúram on DeveloperWorks site in November 2015.
In this article we introduce the concept of multi-temperature data. In general, more recent data in the Cúram application database is more likely than older data to be accessed by queries, batch jobs, deferred process, or to be updated. Therefore, such data is called hot. Analytical data, where data is less likely to be updated, is referred to as warm data in this article. Not all analytics require real-time data. There are classes of analytics where latency in the analytic view is acceptable or even preferred. Analytical data, where data latency is acceptable, is referred to as warm data in this article.
With warm data, you have possibility of a classic software trade-off. You can trade warm data in a cohort of analytics, to reduce the analytic workload on the application database. This article provides a tool that allows you to read warm Cúram data, so that you can create analytics and remove workload from the live Cúram application database.
Read the article Reading Multitemperature Data With SPMP Analytics
Download the sample code.