Google's Caffeine Update Running Off Google's New File System (GFS)

Aug 19, 2009 • 9:26 am | comments (6) by twitter Google+ | Filed Under Google Search Engine

Shortly after Google announced the Caffeine preview update The Register published an article named Google File System II. The two are actually mostly unrelated, in that the caffeine update runs off the second Google File System, like Microsoft Word runs off Windows.

But Matt Cutts confirmed in a recent interview with The Register that Google's Caffeine infrastructure, does indeed run off the second Google File System. The article said, "Matt Cutts confirms that the company's new Caffeine search infrastructure is built atop a complete overhaul of the company's custom-built Google File System."

"There are a lot of technologies that are under the hood within Caffeine, and one of the things that Caffeine relies on is next-generation storage," he says. "Caffeine certainly does make use of the so-called GFS2."

How does this impact SEOs? Well, honestly, it doesn't. Caffeine will have an impact, but it running on GFSII really doesn't matter technically to SEOs.

Forum discussion at WebmasterWorld.

Previous story: Google Again Able To Charge Reserve Bank of India (RBI) Cards


Phill Midwinter

08/19/2009 03:53 pm

I disagree that this isn't relevant to SEOs - a new file system means a new index, changing the structure of an index means you can incorporate all kinds of things that weren't previously possible. I'm also slightly nonplussed after asking Mr Cutts on Twitter: @mattcutts Is Google's new file system ( the reason behind new the index created for Caffeine? mattcutts@phillmidwinter nope, but that's exciting infrastructure too :) I can see he read some leniency into his reply, but i feel like he smudged the truth a little. Not that he's responsible to me or any other seo!

Barry Schwartz

08/19/2009 03:58 pm

I am not understanding the issue with Matt's reply. Caffeine doesn't have to be related to the new GFS, it just happens to run on it. And no, the new file system does not mean a new index. Caffeine means a new way of indexing and again, technically, it can run on the old file system. That is my point.

Phill Midwinter

08/19/2009 04:07 pm

He didn't mention that it was running on GFS2 at all, he had the twitter characters left :( You're quite right it doesn't mean it has to have a new index, but it is a new index... we've been told that - and wouldn't it make sense to construct it differently given GFSII and a fresh start?

Phill Midwinter

08/19/2009 04:10 pm

Also - as it's on GFS2 - does this mean that the speed improvements that people have been talking about (and potentially the increased index size - if you can run things faster you can maintain a larger fresher index more effectively) could be all due to the new file system?

Barry Schwartz

08/19/2009 04:13 pm

I doubt they are "all due to the new file system," I would guess some are due to that and some are due to the infrastructure changes to the indexer. And yes, the article above does say that Matt confirmed caffeine is running on GFSII.

Michael Martinez

08/19/2009 07:18 pm

They can run an old index on top of a new file system. However, what it could mean for the SEO community is that its contributions to faster turnaround on indexing help provide better feedback on how well our optimization works.

blog comments powered by Disqus