Google's Datacenters on Punch Cards
There are several discussions going on about Google Data Takeout, and of the total demand this puts on Google. Data are not readily available, though we've got a few broad outlines.
xkcd's Randall Munroe tried to pin down Google's total data archive about six years ago, in 2013, and came up with about 10 exabytes of disk-equivalent storage, and likely another 5 EB of tape. The doubling rate is roughly every 2 years, so that's probably somewhere in the ballpark of 80 EB and 40 EB now, using very rough numbers.
In ballparking total Google+ data storage, I'm relying again on Stone Temple Consulting's March 2015 public user activity estimates (based on my own earlier estimates, and methodology), which suggest a core of about 50,000 highly active users, and another 1-30 million or so who are occasionally active. If the average of 50 million users has 100 MB of G+ data, that's about 5 petabytes of data, and I suspect the total is smaller, perhaps 10x so. But we're likely looking at 0.5 - 5 PB, or on the order of, in my estimate.
Enough that automating an initial archival on behalf of users, to reside for 12 months after G+ sunset, could be quite reasonable.
Though not delivered as punch cards.
https://what-if.xkcd.com/63/
https://what-if.xkcd.com/63/
There are several discussions going on about Google Data Takeout, and of the total demand this puts on Google. Data are not readily available, though we've got a few broad outlines.
xkcd's Randall Munroe tried to pin down Google's total data archive about six years ago, in 2013, and came up with about 10 exabytes of disk-equivalent storage, and likely another 5 EB of tape. The doubling rate is roughly every 2 years, so that's probably somewhere in the ballpark of 80 EB and 40 EB now, using very rough numbers.
In ballparking total Google+ data storage, I'm relying again on Stone Temple Consulting's March 2015 public user activity estimates (based on my own earlier estimates, and methodology), which suggest a core of about 50,000 highly active users, and another 1-30 million or so who are occasionally active. If the average of 50 million users has 100 MB of G+ data, that's about 5 petabytes of data, and I suspect the total is smaller, perhaps 10x so. But we're likely looking at 0.5 - 5 PB, or on the order of, in my estimate.
Enough that automating an initial archival on behalf of users, to reside for 12 months after G+ sunset, could be quite reasonable.
Though not delivered as punch cards.
https://what-if.xkcd.com/63/
https://what-if.xkcd.com/63/
Comments
Post a Comment
New comments on this blog are moderated. If you do not have a Google identity, you are welcome to post anonymously. Your comments will appear here after they have been reviewed. Comments with vulgarity will be rejected.