3 ms·

"The main problem with Data Furnaces, of course, is physical security." *The main problem with Data Furnaces, of course, is summer.

seles 15y ago
Microsoft, here is a marketing tip: release ideas like this in the winter!...
aridiculous 15y ago
Perhaps the enclosure could be separated from the room (much like an underground water boiler) and a valve could be used to switch between seasons. My main problem with the study is that they probably didn't factor in overhead costs. Once the finances make sense — if they ever do — you can probably bet on some kind of success. Corporations are one track minded and will follow the bottom line. The tricky part will be changing consumer behavior.
raganwald 15y ago
Imagine, if you will, that you have a computing furnace. When you turn it on, it makes itself available to the cloud. You receive credit for the cycles used, enough to pay for the electricity it consumes. On hot days, you turn it off. This seems very much like having your own windmill connected to the grid. In the case of computing furnaces, there would be more capacity in the winter and at night. But at some point when bandwidth and hardware become cheap enough, these would be feasible even if they don't run 24/7/365.
- kevinchen 15y ago
  The problem is, what if somebody stores their data in your server, and then you decide to turn it off? Will the data have to be automatically mirrored on several other servers (inefficient) or will it have to be mirrored when you want to shut off your computing furnace (slow)?
  - raganwald 15y ago
    In my future, all data is distributed P2P like torrents, as is all computation. The only correct architecture for a server is that to turn it off, you simply turn it off :-) But I was thinking more about computation than data, even though the article author is thinking of data. So EC2, not S3.

3 ms·

"The main problem with Data Furnaces, of course, is physical security." *The main problem with Data Furnaces, of course, is summer.

by Thieum22 15y ago

seles 15y ago
Microsoft, here is a marketing tip: release ideas like this in the winter!...
aridiculous 15y ago
Perhaps the enclosure could be separated from the room (much like an underground water boiler) and a valve could be used to switch between seasons. My main problem with the study is that they probably didn't factor in overhead costs. Once the finances make sense — if they ever do — you can probably bet on some kind of success. Corporations are one track minded and will follow the bottom line. The tricky part will be changing consumer behavior.
raganwald 15y ago
Imagine, if you will, that you have a computing furnace. When you turn it on, it makes itself available to the cloud. You receive credit for the cycles used, enough to pay for the electricity it consumes. On hot days, you turn it off. This seems very much like having your own windmill connected to the grid. In the case of computing furnaces, there would be more capacity in the winter and at night. But at some point when bandwidth and hardware become cheap enough, these would be feasible even if they don't run 24/7/365.
- kevinchen 15y ago
  The problem is, what if somebody stores their data in your server, and then you decide to turn it off? Will the data have to be automatically mirrored on several other servers (inefficient) or will it have to be mirrored when you want to shut off your computing furnace (slow)?
  - raganwald 15y ago
    In my future, all data is distributed P2P like torrents, as is all computation. The only correct architecture for a server is that to turn it off, you simply turn it off :-) But I was thinking more about computation than data, even though the article author is thinking of data. So EC2, not S3.