• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Sandforce controller + spindle-drive = much much faster spindle drive?

Idontcare

Elite Member
I was wondering if the IP features (data dedupication, compression, etc) which set Sandforce apart from prior gen SSD controllers might be able to bring value to the spindle-drive market as well.

Wouldn't data deduplication and on-the-fly compression, i.e. anything that reduces write amplification, boost the throughput of a spindle-drive? Latency would not be improved but wouldn't bandwidth see a substantial improvement?
 
market as well.

Wouldn't data deduplication and on-the-fly compression, i.e. anything that reduces write amplification, boost the throughput of a spindle-drive? Latency would not be improved but wouldn't bandwidth see a substantial improvement?

I'm sure it can as I've seen speeds far exceeding what a USB port is capable of restoring compressed data!
 
I'd thinke yes, but those controllers are rather expensive aren't they? Wouldn't that diminish one of the biggest advantages of HDDs (prize) - since you can get 1.5TB for 70€ here, even only a moderate sum would increase the prize a good bit.
 
I was wondering if the IP features (data dedupication, compression, etc) which set Sandforce apart from prior gen SSD controllers might be able to bring value to the spindle-drive market as well.

Wouldn't data deduplication and on-the-fly compression, i.e. anything that reduces write amplification, boost the throughput of a spindle-drive? Latency would not be improved but wouldn't bandwidth see a substantial improvement?

They are actually used... In ZFS you can enable data dedup and on the fly compression.
compression increases performance, currently dedup greatly decreases performance since it is done on the fly, and with large datasets you run out of ram, have to use the hdd to store the rest of the dedup table and performance is atrocious when doing so. (my 5x750GB array had tons of thrashing and huge slowdowns with dedup on)
 
Back
Top