Blind spot - the trouble with optical drives
Why so maligned?
Optical drives should be popular. They hold lots of gigabytes, can be intrinsically WORM and hold data for a guaranteed 30 years or more without the periodic refreshes needed by tape. But they are simply not used as a general data archiving product by business, having become niche products in areas such as medical image recording.
Why might this be?
They are slow. This should not be a problem, as tape is slow. Both are slow compared to disk and whoever wanted fast access from an archive medium anyway?
They are small. Again, this is compared to tape cartridges. But why on earth should this be a factor?
They hold less than tape. LTO-4 holds 800GB of raw data. UDO-2 holds 60GB. That means you need 13 UDO-2 disks for every LTO-4 tape cartridge.
They don't come in cartridges, or, at least, CDs and DVDs don't, coming in insubstantial jewel cases instead. This means that they are totally unsuited to automation devices as the robotics could damage the disks when picking them up and moving through the innards of the automation device, typically known as a juke box.
There is no multi-supplier standard for a cartridge-enclosed optical disk. Plasmon's UDO format is the premier, indeed pretty much the only, enterprise-grade optical disk but it is a single source technology. The previous generation magneto-optical (MO) disks and drives were made by several suppliers but have faded away from the market.
It is possible that, as the multi-sourced and therefore open LTO tape format has displaced all other competing tape formats in its market, there needs to be an open enterprise optical standard before optical storage becomes popular.
Plasmon found it hard going establishing its UDO technology with tough and onerous captial -raising and company re-organisations necessary before it found a way to sell it by presenting as just one media part of an overall archive product including software and multiple media types.
The technology is reliable and has a good capacity roadmap but that has not been enough on its own.
This suggests that other proprietary optical technologies will fail unless they offer overwhelming technology advantages, such as a combination of high capacity, fast I/O, 30-year plus reliability, cartridge encasement, and affordability. It is unlikely that any one supplier has the capability of providing all five of these requirements plus a sales channel capability.
This brings us to look the situation of InPhase's holographic disk which offers 300GB capacity, more than the current UDO-2, but has been held back by development problems. It also has a quite expensive drive with $12,000 being mentioned. Not surprisingly it is being marketed as a movie storage mechanism for prefessional movie archive needs.
Call/Recall is developing a 1TB optical disk that would be compatible with Blu-ray. Blu-ray compatibility would hopefully provide an extended market. There is no good information about productisation schedules and pricing.
Pioneer has just announced a 400GB optical technology that could have Blu-ray compatibility. There is no information on productisation schedules and pricing either.
Enterprise optical disk use is held back by, in my estimation, six factors:
- Lack of multi-vendor standards
- Lack of compelling capacity
- Lack of I/O speed
- Lack of reliability
- Lack of compelling durability advantages
- Lack of affordability, given that DVDs or Blu-ray are simply not enterprise-class
Until a vendor builds an enterprise-class optical drive and disk technology that ticks most or all of these boxes there is little probability that a substantial enterprise optical disk market will develop, however many layers there are in a gee-whiz technology coup.
Copyright © Blocks and Files
Sponsored: DevOps and continuous delivery