(via Steve Dekorte)
Fusion io's flash storage card. Neat.
...the cards will start at 80 GB and will scale to 320 and 640 GB next year. By the end of 2008, Fusion io also hopes to roll out a 1.2 TB card...Five years from now will be fun, running not-so-little data centers in a pizza box....the card has 160 parallel pipelines that can read data at 800 megabytes per second and write at 600 MB/sec. He even proved it by running a Linux drive I/O benchmark. But for large corporations running busy databases, operations per second is a much more important number than bandwidth.
Flynn set the benchmark for the worst case scenario by using small 4K blocks and then streaming eight simultaneous 1 GB reads and writes. In that test, the ioDrive clocked in at 100,000 operations per second. “That would have just thrashed a regular hard drive,” said Flynn.
How are you going to justify running your operations on a mainframe in 2012?
Moore's Law is changing the hardware landscape orders of magnitude more quickly than the software community can track. We have not even grasped the difference between today and tomorrow because we're still way back before yesterday in the way we think about software.
How long before someone gets rid of that artificial disk driver sitting between the processor/caches/memory and the "disk"?
2 comments:
We have not even grasped the difference between today and tomorrow because we're still way back before yesterday in the way we think about software.
That just means more low fruit for us to pick.
Last time I looked at mounting flash drives in Linux, the prevailing wisdom was that you absolutely should turn off atime (o=noatime) when mounting a flash volume because leaving it on would drastically reduce the lifetime of the flash media. At the time, that was because you had a finite number of write operations on flash media and updating file atimes wasn't a smart use of those limited writes. Wonder if they have fixed that yet. Seems like they will need to before we can really consider using these large volumes for things like database storage.
I now mount all my Linux volumes without atime, provides a nice little speed bump, could help some of those Erlang bottlenecks maybe :)
Post a Comment