Tatiana Bunto
Analyst, HFLabs
Sorry, this entry is only available in Russian.
Analytics for 100 million data. Short lesson for system integrators
October 13, 13:10
Room II|II зал
Discuss the presentation
In enterprise systems, data has been accumulating for years. And it seems a simple task-to configure the integration of the database sources in the database of your system. It would seem, what problems can be? Check the model of the system, do mapping. The catch is that everyone lies! There is no perfect data! And our task is to cut off all unnecessary and not to lose valuable. And for this you need to carry out analysis.
Cherry on the cake are the volumes. Overloading 100mn of data is not equal to overloading 10mn. For 100 million, it is necessary to take into account the features of the model, the future purpose of the system and its use scenarios,
At what stages of system integration is it necessary to analyze the data? How to check why? And most importantly-what to do with the results?
I will share my experience, interesting cases and examples. I’ll give you tips on data analysis when designing system integration.
The report will be of interest to everyone who is interested in the quality of data in the integration of systems, wants to put in order the database of its system, as well as those who make project plans and do not understand why spend time on such research.
