One question before I dive in!

Hi there! My Dropbox recently got deleted and I lost 2 years worth of video and photo work. Whatever it was all obsolete because this will be my best year but needless to say I’m pretty p!ssed and looking for an alternative. I like Storj and looking into using it for store Video and Photo work and send share links to clients, etc.

What mechanism safeguards files from being lost? If someone goes offline or their node gets wiped would part of my data be lost forever?

1 Like

I am a node operator. Nodes never store a complete file but only pieces that are encrypted and erasure coded. Erasure coding is the key here because this makes sure that you need only a few pieces to reconstruct your file again.
Currently the ratio is 29/80. This basically means that only 29 pieces of 80 are required to reconstruct your file (segment to be correct). These pieces are distributed to nodes all around the world.
No node stores more than 1 piece of a segment. So even if some nodes fail or go offline, there are plenty of other nodes available holding enough pieces for reconstruction.
In the background there is also a repair service. If nodes are offline too long and the number of pieces of a file is getting low, the repair service recreates pieces and distributes them so that there are always enough pieces available.
Storj claims they have never lost a file until today. So this process seems to work.


This is a wrong approach.

Regardless of what solution you use, you need to have backups. Onsite backups, offsite backups, disaster recovery backups, cold storage backups.

There are many more scenarios that can result in data loss, outside of a vendor failure.


As explained above, Storj uses erasure.encoding and replication to ensure files are never lost. I just wanted to mention that this happens seamlessly behind the scenes and you don’t need to do anything or reconstruct anything yourself. You would upload and download files like normal. Storj ensures there are enough pieces available and creates new ones as needed so that there are always enough to recover your data even if a significsnt portion of nodes were to go offline.


As @arrogantrabbit alluded to, it’s absolutely crucial that you have backups and a disaster recovery plan if your data is essential for your business.
May I ask how it was that Dropbox lost your data? It seems unlikely (but certainly not impossible) that it was a technical fault at their end, in which case it might have been user error and no platform will help you against those…


Accidental deletion is the most common form of data loss - and only solved by having regular backups.


I believe that having a versioning: Object Versioning Closed Beta could help a little bit.

1 Like