* Posts by Computurd

8 publicly visible posts • joined 19 Mar 2015

Want to use 3D XPoint DIMMs with Intel Purley Xeons? Wait a couple of years for second-gen

Computurd

the ceo is speaking about on=package 3DXP, not dimms. Dimms will work with anything with proper code support, it dosnt require a hardware change to the CPU.

Mangstor, Mellanox flash rig crowned 'fastest in the lab'... for RAID-0

Computurd

Interesting.

NVMdurance extends flash life tenfold

Computurd

Dynamically adjusting the voltages of the individual cells is hardly new, or even exciting, tech. SMART Storage and STEC began doing that in 2011. Now there are a few more hooks exposed, such as in Micron Fortis Flash, but others are also exploiting this technique as well. Again, nothing new.

Explain the "new" part again? I'm having a hard time understanding what is being done that is not already done elsewhere. From the descriptions in the article, and on the NVMdurance site, none of this is new.

Computurd

This is merely DSP (Digital Signal Processing), which is a staple in all of the currently shipping SSDs. Nearly every SSD has this technology.

Hold on a sec. When did HDDs get SSD-style workload rate limits?

Computurd

These specifications are nothing new - they have been on HDDs for years. It isnt due to shingling, only one of the models even has shingles, the Seagate 8TB, the rest do not.

A comparison of DWPD (Drive Writes Per Day) is frivolous with an 8TB volume.

Pure's mass disk drive killer lays out plans for flash hegemony

Computurd

Deduplication and compression also apply to HDD storage as well, an oft-overlooked fact, especially when looking at cold storage, which is the vast majority of data. Sure, disks cannot absorb the performance hit quite as easily, but you are burning CPU cycles for all that dedupe and compression anyway no matter the medium, and for cold data the performance overhead is tenable.

Flash banishes the spectre of the unrecoverable data error

Computurd

FUD

The problem is this. Modern RAID controllers do not scuttle the entire array in the case of another error during rebuild. That is FUD. they havent done this for ten years. You lose that bit of data, sure, but the rebuild goes on.

The storage is alive? Flash lives longer than expected – report

Computurd

These SSDs were never powered off not once. that is the fail in this testing. Endurance is rated for data retention with the power off. Several of the guys who did the initial testing (that this testing is based off of) had SSDs 'die' after one second of power off after they had reached their warrantied writes. If you removed power from any of those SSDs, period, even halfway through that test every single bit of data would be gone.