Messages in this thread | | | Date | Mon, 22 Mar 1999 09:28:22 -0800 | From | Guy Sotomayor <> | Subject | Re: [OFFTOPIC] Re: disk head scheduling |
| |
> On 22 Mar 1999, David Wragg wrote: > >> Why SRAM? The nice thing about SRAM is its low latencies. If you're >> slurping a whole track off a disk, latencies shouldn't be an issue; >> you might need lots of bandwidth, but you can get high bandwidth from >> DRAM. >> > > Because dynamic RAM needs refresh. Refresh takes power. You need > to keep the data without power, i.e., the bias from a small battery > that is essentially shelf-life. > What are you talking about? Seriously! What drives have batteries on them to keep their RAM alive? This is a joke right? I haven't seen any drives that count on the persistence of the RAM to maintain data integrity (this doesn't mean they don't - just unlikely that they'd be mass market stuff). Remember there is a lot of stored energy in those spinning discs. If a drive vendor really wanted to maintain data integrity, they'd use the spinning inertia of the platters to reverse power the motor to allow a write in progress to complete before loosing power. All the drives indicate is that any command they mark as complete is (ie data on the disk). Anything else is up to the file system.
TTFN - Guy
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |