In this world of virtualization and clouds this piece of news about Spectralogic is quite interesting.
Spectralogic posted a +9% year on year sales increase (and an impressive +14% about enterprise tape libraries sales!). I expect SpectraLogic will also announce something new next week (an integrated disk/tape product?) and I’ll be there to have the news straight from the horse’s mouth.
I recently wrote a couple of blogs on the role of tapes and the evolution of disks (especially in the field of cold storage) and next week I’ll have access to more information to get more of a complete picture.

Tape yes

Tape is still the media with the lowest TCO for storing any form of cold data (long term backups, archiving, etc.). Especially when you have data that are written once and never retrieved it’s almost perfect.
Tapes can also be stored outside the libraries in some cases to improve space utilization or security at a low cost.

Tape no

spectra_t_finityThe problem with tapes are well known too. It’s a sequential media in a world of random requests.
You can store a tape for 30 years or so but who will be able to read it then? So migrations come every now and then (it depends on compatibility between different tape drive generations, tapes consolidation, and so on).
In a few words, it’s complicated and there are some hidden costs that come up during the life of a tape.

Bottom line

Tape is here and it’s here to stay (I’m also a bit surprised by the +6% y/y on small libraries sales, that’s in counter trend if compared with other small libs players).
On the other hand tape needs to evolve to become more usable and aligned with the needs of modern IT architectures. First attempts, like LTFS for example, look half-backed to me and users ask for more: they want cheaper and easy to use storage… and, please, don’t think about traditional HSM, it isn’t sexy enough (don’t ask me why!).

WS-ENG

Disclaimer: I was invited to Spectra Logic event and they paid for travel and accommodation, I have not been compensated for my time and am not obliged to blog. Furthermore, the content is not reviewed, approved or published by any other person than the Juku team.