Arbor has now provided more detail on the astounding explosion of Google's Internet presence, which as of last summer it estimates as accounting for peak rates of 10 percent of all Internet inter-domain Internet traffic it sees travelling through its servers.
Between June 2007 and a year later, the average traffic percentage grew from around 1 percent to around 2.5 percent; by last summer the percentage was a minimum of 5 percent and growing.
The main reason was Google's acquisition of YouTube in 2007, which consumes huge volumes of video traffic, the application that almost on its own is driving capacity growth at the peer network level.
As significant as their sheer number is how all these Google-related packets move across the Internet. In mid-2007, Google used third-party "transit" (i.e other networks) for a large percentage of its Internet traffic. By this February, Arbor reckons that over 60 percent of Google's traffic was being channelled through direct interconnects that link its massive data centres to one another.
To put this in less technical terms, Google and the customers using its services are not so much using the Internet as Google's own private corner of it, a peered network within a wider Internetwork.
Arbor's Labovitz reminds us that Google has apparently spent the last year installing Google Global Cache Servers (GGCs) in as many as half of all third-party consumer networks in the US and Europe, which extends the edge of its network into even more data centres.
"Unlike most global carriers, Google's backbone does not deliver traffic on behalf of millions of subscribers nor thousands of regional networks and large enterprises. Google's infrastructure supports, well, only Google," comments Labovitz.
Famous for its search, e-mail and YouTube video sharing, Google has quietly and relentlessly turned itself into the first super-carrier of the Internet era.