This article is more than 1 year old

Violin: Don't weep for Tier 1 storage... it'll soon be flashtastic

CEO: Scale-out hyper-converged systems? Pah!

Three dimensionality

Layering planar NAND cells will add capacity to NAND dies by making them thicker, by building (as it were) a block of apartments instead of single story buildings with the same footprint. This can be done with existing MLC flash, the two-bits-per-cell stuff, which is inherently faster and has a much longer working life. Also, with so much (3D) space to play about in, theoretically, you can use an older, and larger cell geometry and get even longer endurance while still increasing chip capacity.

DeNuccio says that Violin has expanded from having a point product to having a full set of flash products providing a tier 1 storage alternative. "Ultimately, we can see a move by the market to all-flash," he says, arguing most visionaries are only seeing the potential for flash as a tier 1 storage medium with disk providing the chap and deep bulk storage.

"3D will make the difference ... [there] will be an enormous difference in cost [and] it could change the players." He sees three main operators for flash as Samsung, SanDisk and Toshiba. Not Micron, although it "has a lot of DRAM equipment it could transfer to NAND".

Violin will benefit from 3D NAND: "[With] products in the first half of next year ... we'll distance ourselves dramatically from anyone else using SSDs [and] the next-generation system will be a leap forward in hardware architecture."

He argues SSDs are over-provisioned by 50 per cent and Violin can do much better than that with its VIMM technology. "We have visibility into the entire NAND stack. SSD users don't." Both all-flash array suppliers and server-side suppliers using SSDs will be at a technology disadvantage because of this.

Waves of change

DeNuccio thinks tier 1 storage's move to all-flash will happen very fast, with "an enormous effect on existing vendors".

DeNuccio also talks of waves of change hitting the industry, with "EMC in play. HP splitting up, [and] cloud, virtualisation, and flash are all happening at the same time. It's very disruptive".

He predicts business IT use will bifurcate, with SME users going to the public cloud and larger enterprises using private clouds — bursting to the public cloud where necessary and then reverting to their private clouds when the burst need us over.

It makes us think that the old question — what storage startup can emerge from the pack to become a platform company, as NetApp did many years ago — is no longer relevant. It's the wrong question.

Unlike NetApp, which grew into a full disk storage protocol and architecture company – NAS, iSCSI, SAN FC, single and clustered arrays – the storage world Violin is growing into is not disk-based any more. The storage game rules have changed and software can add the pieces Violin (and other flash array startups) need without them building head-to-head products to compete with the mainstream vendors’ legacy product set.

Can Violin ride the change waves and emerge back on top? It’s too early to tell, but the company is in goodish financial shape, re-energised in product development and better organised in its channel operations. ®

More about

TIP US OFF

Send us news


Other stories you might like