3 ms·
For http://lanyrd.com/ http://lanyrd.com/ we treat Solr as a denormalisation layer for our MySQL database - there are plenty of queries that are pretty gnarly i
by simonw 15y ago
For http://lanyrd.com/ http://lanyrd.com/ we treat Solr as a denormalisation layer for our MySQL database - there are plenty of queries that are pretty gnarly in SQL which Solr handles with ease.
Aa good example is our http://lanyrd.com/calendar/ http://lanyrd.com/calendar/ page which shows events your Twitter contacts are attending in the future - this was horrible as a SQL query but works great as a Solr search, especially as scaling Solr read-slaves horizontally is pretty simple.
Holding denormalised data in a separate data store entirely also makes it easier to spot bugs where denormalised data isn't being correctly updated.
- mattdeboard 15y agoWe do the same as well wrt denormalization, though it was not the original intent of the refactor. Nearly every item on our job sites started life at the beginning of the project's lifecycle as a "pretty gnarly" SQL query, but now lives on as a highly performant Solr query. (Admittedly we don't have nearly as many widgets going on as you do!)