Last week, LinkedIn announced a new search architecture which they are calling Galene. Siram Sankar, Principal Staff Engineer at LinkedIn, authored a very detailed blog post explaining the need for Galene, what it is currently doing, and what they plan for Galene to be able to do in the future.
Below are a couple bullets covering the basics of the upgrade:
1) Instant Member Search (the searches conducted at the top of the page) is now able to search the entire LinkedIn database. Previously, instant search only searched for 1st and 2nd tier connections. LinkedIn also claims that searches are now more relevant, faster, and more efficient.
2) LinkedIn will now be able to develop further search functionalities which will be able to explore the entire LinkedIn economic graph. The example searches provided as examples are very similar to Facebook graph searches (not surprising given that Sankar previously worked on search quality/ranking issues for Facebook). See below for the example economic graph searches:
If you’re looking for more technical details, read the entire post from LinkedIn.