0

i'm developing a system that will collect user activities samples (opened a window, scrolls, enter page, leave page, etc.) and i'm looking for the best way to store these samples and query it.

i prefer something smart where i can execute sql-like group by queries (for example give me all the window open events grouped by date and hour), and of course something flexible enough in case i'll need to add columns in the future.

i'm trying to avoid thinking about all the queries i might need and just save an aggregated version of the data by time, since i'd like to do drill-downs. (for example, count all the window open events by date and time, and then see all event in each time-frame, or change it to be by unique userId).

thanks.

PS - i currently use MySql for this task, but the data is expected to grow rapidly. I've experimented with mongoDB as well.

4

1 回答 1

0

我相信 mongoDB 可以是一个很好的解决方案。首先,它旨在保存大数据,并且非常易于使用和扩展(副本集或分片)。表达语言也很扎实。我的意思是它不如 SQL 强大,但仍然足够好。这是一个关于将 SQL 命令映射到 MongoDB的好链接。

还有其他替代方案,但我认为要么它们太复杂,要么它们的表达语言不够强大。

也请查看此链接,它可以帮助您找到适合您的解决方案。

于 2012-06-29T16:07:29.633 回答