Music library (backup) in the cloud?

[Qhwoeprktiyns]

18 minutes ago, AudioDoctor said:

I don't actually use mine to listen to music, just to backup. I could however.

 

That is indeed how many people do it today - and the subject of the thread.

In the coming years, we will probably see improved cloud based players, and the situation may be reversed, with "local" storage being purely for backup purposes, if at all.

[bodiebill]

2 hours ago, hopkins said:

 

Sorry, i had not taken the time to explain in detail.

 

Unfortunately what I am doing is a "custom" solution and I do not really see how it could be extended to others without investing significant time and effort into it. So i don't think a seperate thread is required. I'll explain here briefly.

 

I developped everything from scratch. My program scans automatically my folders. 

I store  album covers and associated artwork/pdf  on Google Cloud Storage. The "tags" and comments on my albums are stored in a PostgresQL database hosted on Heroku. 

 

The tracks themselves are for the time being only stored locally, but i plan on storing them as well on Google Cloud Storage, selectively for the time being, to allow me to  play them remotely.  For the time being, I play my files through LMS by sending the URLs of the files from my app. Works well, but it ain't Roon... 

 

That's about it

 

I put a few examples here: 

The example illustrate why i did all this. They are individual albums. I also have an "album grid" to browse my entire collection, but it is not finalized. I plan on adding artist bios/pictures as well - it all takes time. I spend more time "collecting" information on my ever-growing collection than on actually developping the solution. 

 

By the way, i am the author of this thread: https://community.roonlabs.com/t/bye-roon-why-i-am-not-renewing-my-subscription/67012

 

I followed through on my ideas and am very happy to have done so  I mostly listen to Jazz, and as a Jazz "collector", there are in my opinion many aspects which are just not handled by any music application today. Unfortunately,  there is no "collaborative" database that address these requirements. A shame, but that's the way it is.

 

Thanks @hopkins, sounds like an intricate custom setup!

[Jud]

45 minutes ago, AudioDoctor said:

I don't actually use mine to listen to music, just to backup. I could however.

 

I've got the backup in the cloud and local, and remote music listening enabled at home with my desktop as the server. Will likely build a NAS one of these days for the remote listening and local backup/storage.

[Qhwoeprktiyns]

33 minutes ago, bodiebill said:

 

Thanks @hopkins, sounds like an intricate custom setup!

 

It is  intricate in its "design", but it remains simple in its use: scan your folders,  browse, play, and edit your albums in a web page. 

 

For playback, I do plan on making it simpler with cloud based playback (have to check the cost of having all my collection on Google Cloud Storage or Amazon AWS), but as you may know audio playback in itself is a complicated topic... There are too many protocols (Web Audio, UPNP, LMS, casting, etc.. ), and some protocols/solutions are really poorly designed (Web Audio to start with...).

 

Back to cloud storage backup - sorry for the digressions...

 

[AudioDoctor]

4 hours ago, Jud said:

 

I've got the backup in the cloud and local, and remote music listening enabled at home with my desktop as the server. Will likely build a NAS one of these days for the remote listening and local backup/storage.

I have 1 backup here and I have it mirrored on my brothers NAS as well. I’m not tech enough, nor willing to learn, to build one. Plus the iosafe/ Synology NAS is as disaster proof as one can get. 

[Jud]

13 hours ago, AudioDoctor said:

I have 1 backup here and I have it mirrored on my brothers NAS as well. I’m not tech enough, nor willing to learn, to build one. Plus the iosafe/ Synology NAS is as disaster proof as one can get. 

 

Not trying to persuade you because you already have a preferred solution, but any kind of computer building, including a NAS, is a very prosaic thing. It's pretty obvious which pieces click or screw together where. Putting together a kid's bike might be more difficult.

[The Computer Audiophile]

2 minutes ago, Jud said:

Putting together a kid's bike might be more difficult.

"Ruined" my wife's bike last summer when I had to replace the rear wheel's tire. I took the rear derailleur off and it was all over for me. 

 

NAS? Done it in my sleep many times.

[jabbr]

19 hours ago, hopkins said:

 

Cloud storage/apps arise happening everywhere else... Corporations are moving to cloud based solutions.

 

Having to run a home server 24/7 to listen to your music files (at home or remotely) seems "archaic" to me. A chromebook, a backup solution (local or remote) - that's all we should need.

The two things are very very different.

 

”Cloud” backup is just that. Typically you pay less for less frequent access. 
 

A home server is used to expose your data in different ways:

1) as SMB files

2) as DLNA 

3) as Roon/RAAT

4) as HQPlayer streams

 

or combinations of above. In any case if you care to do upsampling, it’s a processor intensive activity for the home server.

 

A NAT is just a computer with an Ethernet port and typically removable disc drives. Linux eg Ubuntu is free and contains everything you need to have a fully functional NAT. It will also run on whatever old computer you have laying around  

[jabbr]

21 hours ago, hopkins said:

Sorry, i had not taken the time to explain in detail.

 

Unfortunately what I am doing is a "custom" solution and I do not really see how it could be extended to others without investing significant time and effort into it. So i don't think a seperate thread is required. I'll explain here briefly.

ok nice... Amazon Music 

[Jud]

19 minutes ago, jabbr said:

The two things are very very different

 

I went through this a while back with @AudioDoctor. As I recall, his home server is duplicated at his brother's, reasonably fire and flood-resistant, and bolted to concrete to prevent theft. So it may not be Google but it's reasonable backup. 🙂

[AudioDoctor]

1 hour ago, Jud said:

 

Not trying to persuade you because you already have a preferred solution, but any kind of computer building, including a NAS, is a very prosaic thing. It's pretty obvious which pieces click or screw together where. Putting together a kid's bike might be more difficult.

 

You are, of course, correct Jud, its more the software and such that I don't feel like getting into where the Synology is effectively plug and play for me and was easy to figure out and learn with help from a local IT expert.

 

9 minutes ago, Jud said:

 

I went through this a while back with @AudioDoctor. As I recall, his home server is duplicated at his brother's, reasonably fire and flood-resistant, and bolted to concrete to prevent theft. So it may not be Google but it's reasonable backup. 🙂

 

And then there is this as well, it is fire and disaster proof (somewhat) and was easily duplicated. I am a big proponent of the KISS principle. Keep It Simple, Stupid.

[austinpop]

Another data point here for peer to peer backup. In my case, I have a friend across town, and we each own a primary NAS in our own house, and a remote backup NAS in friend's house. Both of us upgraded our NASes and used the previous one as the backup target. These are all Synology, and we just use HyperBackup.

 

The nice thing about this approach is the initial backup of TBs of data can be done over your own local home Gb network. For both of us, the initial backup at 1GBps speed took a day or two at most. After the initial backup is done, deploy the backup unit at friend's house, and do the one-time router setup to open ports etc. Synology has tools for this.

 

Of course this is in the context of our ISP speeds, which sadly is limited to 300/30 Mbps.

 

We've been doing the above for over 3 years. I haven't looked at cloud backup solutions since then. Have any evolved to the point of allowing you to send in an initial seed on a disk drive?

[ray-dude]

I used to use CrashPlan, but they dropped consumer accounts.  Now I use Resilio (private bit torrent), and have a hard drive hanging off a friends system (I have their hard drives hanging off mine as well). Resilio basically keeps directories in sync via bit torrent (yes, back to the Napster future

 

we have fiber to the home with no data caps, so a lot cheaper than any cloud service (aside from music, DVD and BluRay rips take a LOT of space)

[jabbr]

14 minutes ago, austinpop said:

Another data point here for peer to peer backup. In my case, I have a friend across town, and we each own a primary NAS in our own house, and a remote backup NAS in friend's house. Both of us upgraded our NASes and used the previous one as the backup target. These are all Synology, and we just use HyperBackup.

 

The nice thing about this approach is the initial backup of TBs of data can be done over your own local home Gb network. For both of us, the initial backup at 1GBps speed took a day or two at most. After the initial backup is done, deploy the backup unit at friend's house, and do the one-time router setup to open ports etc. Synology has tools for this.

 

No problem, of course ZFS (designed by Sun, now Oracle, and now freely incorporated into Ubuntu) allows you to do incremental remote backup. That and daily snapshots.

 

My preference for ZFS is that its machine and OS independent, so I have drives from my old Solaris NAS that I was able to simply plug into my Linux NAS. 

 

The newer Synology NAS (or perhaps higher end) use the BTRFS file system which is similar to ZFS and now has adequate testing. So you should be able to reinstall an array onto Linux and recover the data if your system were ever to have a hardware failure 😳 

 

Now for backup, I use Jottacloud  It runs in the background.

 

[ericuco]

I just searched web on cloud backup and found that iDrive is running a deal for $3.48 for one year with 5TB storage through Tom’s Guide.

 

https://www.tomsguide.com/us/best-cloud-backup,review-2678.html 

[bodiebill]

This looks interesting:

https://www.cubbit.io/

Peer ('swarm') based cloud storage.

Any views?

[The Computer Audiophile]

1 hour ago, bodiebill said:

This looks interesting:

https://www.cubbit.io/

Peer ('swarm') based cloud storage.

Any views?

I looked for a couple minutes and couldn’t figure out how it worked. Perhaps I’m a bit slow. 

[bodiebill]

1 hour ago, The Computer Audiophile said:

I looked for a couple minutes and couldn’t figure out how it worked. Perhaps I’m a bit slow. 

 

As I understand it, in order to have, say, 4TB storage (the maximum per Cubbit box), you need one box and attach 8TB worth of drives. The total network of Cubbit boxes ensures consistent availability of all data, even if you disconnect your own box. It reminds me of the blockchain used for cryptocurrencies, but they argue it is different:

https://help.cubbit.io/hc/en-us/articles/360010794099-Cubbit-vs-blockchain

 

So you invest the box (approx. $300) and the drives, and there are no additional monthly fees.

 

It would be nice if I could attach the Cubbit box directly to my wtfplay audio pc via usb. Not sure yet whether this is possible.