EVERYTHING techy is stop-gap.
Nothing to see here. Move along.
Flash is a necessary waystation as we travel to a single in-memory storage architecture. That's the view from a Fujitsu chief technology officer's office. Dr Joseph Reger, CTO at Fujitsu Technology Solutions, is that office-holder, and – according to him – flash is beset with problems that will become unsolvable. He says we …
...it was called AS/400. Fujitsus claim here is just a deja-vu of the manual for OS/400.
Back then it was of course the other way around (on disk instead of in RAM) but the concept is the same: It is just memory, don't worry where it is, the OS takes care of that..
I seem to recall an ancient ICL system (was it George, much beloved) in the early 1980s that treated all media neutrally. Response time apart, one had no idea if the data came from tape, disc or a bucket of cement.. However, I rather think this chap is addressing a different specification.
George behaved that way, as did VME, although you could drill down into the file properties further if you knew what sort of file you were looking at.
I recall, in an example of the circularity of life in IT, that we spent a few happy months playing with the prototype Electronic Storage System (ESS), a whole gigabyte of persistent solid state memory mounted as a filestore system that at that time cost £multiple millions (I forget exactly how much, but it was betwen 5 and 10 million). When we formatted the duplexed IDMS database that we dropped on there we got both nodes of 39/80 that we using as our playground.
VME was good like that. A file was a file and did the things that a file did, a directory was a directory and did the things that a directory did, a group was a group and so on and so on. Happy days.
What's missing in current digital storage technologies is long-term permanence. My family has some nearly 100 year old photos, some with some evidence of fading. I have 50 year old b&w negatives that will last at least 100 years, maybe 200 without danger of fading (my own careful archival quality processing). The digital info I have is on either burned discs that rely on dyes, which like colour negs and slides fade even in dark storage, and thus will be unreadable in 50 years, even if there is working hardware around then to read it, or on hard drives that also have a limited lifespan.
Archival permanence needs to be addressed.
Interesting that a Tier1 CTO, finally, tells it how it is and identifies some of the fundamental problems flash has. Add to this the appalling performance difference between 10% full and 90% full flash devices and you start to see it isn't the panacea that the flash suppliers claim, oh and neither will phase change either.
With appropriate working practise, digital media will last forever. Appropriate working practise means backups, use of error-correction in archiving, migration to different media as required, etc.. The result will be bit-identical over however long a period you care to name. And with proper backups (including off-site storage), you're good for just about anything except your entire country being nuked. I can be reasonably certain that photos of my son will still be as good as the day they were taken when he reaches 100.
Back when pen and ink was the order of the day, the Catholic church understood perfectly that data preservation was a process and not a property of an item. That's why thousands of scribes spent their lives copying out manuscripts. And because there were so many backup copies made, we still have that data today.
Never mind the entire country ... Good data retention practices should allow the nuking of an entire hemisphere without data loss. I currently have my important data mirrored in Chile, South Africa, New Zealand, Japan, Finland, Scotland, Spain, Duluth, Palo Alto and twice here in Sonoma, And I have full data backups sent "space available mail" to my sister in Burgundy weekly. She returns them the day after reception. They might get ME, but my data will (probably) survive.
Is this kludge over-kill for a home system? Of course. But as a research platform it's (mostly) tax deductible.
I take your points about the benefits of digital copies being exactly identical. I was trying to highlight the value of a media that can be put in a drawer, or some safer storage, and with no further effort be readable in 100 years. You can manage backups for some years but life catches up with you. Trust me.
That negative can be rediscovered. Current digital storage media are highly unlikely to have that advantage.
Biting the hand that feeds IT © 1998–2021