31–40 of 42 posts
I've always wondered this, but why don't we just scale up some data storage device? Like create a massive hard drive, 10 feet across? It has to be more efficient than tons of tiny little drives.
The same probably applies for things like the actuator for the heads, etc.
I've always wondered this, but why don't we just scale up some data storage device? Like create a massive hard drive, 10 feet across? It has to be more efficient than tons of tiny little drives.
1) The head has to move further to read the next sector of interest, which is particularly problematic for fragmented data.
2) It is more difficult to manufacture high data density (AKA high-precision) disks in large formats, as some surface defects are cumulative, and get worse as the disk gets larger.
3) Manufacturing defects which occur pseudo-randomly increase proportionally to the surface area of the disk, so the reject rate increases as a square of the radius.
4) Smaller drives can be spun much faster, allowing for higher data rates, as the centripetal accelerations in the disk are proportional to the square of the radius (and I believe the stresses are proportional to the cube of radius).
For these reasons and many others, HDDs have been moving to smaller and smaller form factors.
Great photos but the default resolution is frustratingly small. Here are the links to the fullsized versions: https://tctechcrunch2011.files.wordpress.com/2016/09/1446837... https://tctechcrunch2011.files.wordpress.com/2016/09/1446872... https://tctechcrunch2011.files.wordpress.com/2016/09/1450070... https://tctechcrunch2011.files.wordpress.com/2016/09/1442533... https://tctechcrunch2011.files.wordpress.com/2016/09/1…
The aesthetic feels like it was meant to be a Bond villain's Arctic lair
I've always wondered this, but why don't we just scale up some data storage device? Like create a massive hard drive, 10 feet across? It has to be more efficient than tons of tiny little drives.
Or else split the I/O into a zillion little independent ports.
"Marketing photos released by Facebook's CEO" is a much more accurate headline, this is clickbait.
Good old Zuck!
I've always wondered this, but why don't we just scale up some data storage device? Like create a massive hard drive, 10 feet across? It has to be more efficient than tons of tiny little drives.
Pretty interesting. I love how bespoke data centers are converging along very similar lines. I'm sure if you were inside a Microsoft, Google, Amazon, or Facebook data center you would recognize the new design touch points. The old data center is dead, long live the new data center. This was the second big thing Google had learned early on: "The equipment is reduced to its basics so it runs cooler. It can also be easi…
I think the lesson came in earlier in the NUMA and MPP machines where they kept trying to cram more stuff on boards that were themselves pluggable into the larger system. This convergence has happened from several directions. It's not all the different from the earlier one that started in the 1960's where they fought cost and inefficiency by getting as few components per box sharing as much as possible. Moores Law te…
That said, there are a number of systems at FB where deleting a crypto key loses the linked data forever--but they still crunch the hard drives just to be really sure. The drive crunching is an incredibly tiny expenditure compared to the massive CapEx and OpEx required to build, stock, and run the datacenters. It's worth it if only for the peace of mind.