A new player in long term data preservation

I follow this domain for quite some time with some companies developing new media or new software even new format iterations on existing media but Cerabyte has created a real curiosity. I can even unveiled that I have participated to some stories in the industry. Long term data archiving could be a motivation but also a frustration and still represents one of the biggest data storage challenges in the industry.

The first one is probably the longevity of the media itself and it appears clearly that passive media are one of the solutions such as tape, optical, microfilm or something new like DNA storage. I have to say that the latter has some interests but is not ready at all and by far to be ready. At the same time, some vendors have invested into special projects like Silica. Then we find data format and the density, we'll cover these later.

One of the things I have to clarify is the notion of cold data and cold storage. Elements around data are based on access frequency and obviously a cold data is data not accessed for some arbitrary amount of time, many of us considers 30 or 90 days, to qualify data as cold. Thus this kind of data needs to be stored on a dedicated storage media not primary storage one for sure but more based on a secondary storage model. In other words continue to store cold data on hot storage I mean primary storage or storage for production. Here the capability to identify, classify, move and manage data across storage tiers is essential and we all remember approaches like HSM or tiering or even migration. We find cold data on primary and secondary storage and the goal of every IT leader is to optimize that associated cost by storing only the data on the right storage and thus align the value of data to the value of storage.

Cold storage belongs to the secondary storage category receiving data from the primary and thus supporting the primary in case of failures, loss, recovery or just needs. This type of storage is built to store what we called inactive, reference or archive data, elements we need to keep but not related to the day to day business. The media is passive by design, it is tape, optical or even MAID - Massive Array of Idle Disks - or other specific media. And in that domain, long term data retention introduces some challenges. On the volume side, secondary storage represents the vast majority of the data volume and a large scale capacity challenge.

It exists a real disconnect between the need to store data over a long period of time and the longevity of the available media to do it imposing technology refreshes and migrations. I wish to insist on the fact that the energy and ESG dimension are key when selecting or developing the media for such utilisations.

Cerabyte, an Austrian company based in Munich, Germany, has started some research and developments around a new media, format, coding and decoding, writing and reading phases to address the long term data retention challenge.

The media is a glass with special coating on it to receive the parallel laser beams in the writing phase. Firm's innovations belongs in all these categories. The team has worked a lot on the media and the writing and recording phases. The coding today is super dense QR code but will be replaced by new dense matrices and they're self protected. The writing is done by a femto laser that can create models very fast and in very small dimensions that guarantees the capacity with a real scalability and density of data.  Reading is achieved by using a microscope so no light emission or very sophisticated sub-systems.

All media are manipulated with a classic tape robotic with similar techniques except the cartridge that stores the glass film is square and not ribbon based or circular like optical platters.

I see here lots of promises and expect to get more details soon even a demo with a prototype. And I wish them to join a future edition of The IT Press Tour.