hey there, we forwarded this to sales/bizdev. it would be great if something cool emerges from that idea submitted. Thank you again for helping us come up with suggestions – it really shows the value of community!!
3 Likes
You need to expand that sales department. There’s a lot of empty Hard Disk space around! 
1 Like
Yes, it would be awesome if Storjlabs could get a foot in the door of that industry.
According to Signiants website:
Built on Signiant’s patented technology, the company’s on-premises software and SaaS solutions move petabytes of high-value data every day between users, applications and systems with proven ease.
This little quote proves that we are talking about a massive amount of data that requires storage and transfer capacity worldwide every day.
1 Like
So… something Storj might be quite well fitted to? Maybe I missed the negative to your point, but the fact that everything is distributed, operating with that aggregated bandwidth too, I would figure it’d be a soft sell over all. Plus, no worry about a single AZ issue- effectively US-C/A/E-W are globally hosted by SNO’s unless they Graceful Exit a particular SAT they don’t want to host for.
From what I understand Videocoin offers CPU power for transcoding video content, not storage.
Yes I believe. Maybe needs additional features to suit the specific requirements of that industry.
i kinda assume that’s what they want to collaborate with storj labs for.
CDN is a combination of those things… i got no idea what videocoin was or is planning.
a similar example could be… “a browser also needs storage, doesn’t mean the programmer / engineer of the browser needs to worry about it…”
could be sort of the same thing… they could be thinking, there are a good deal of interesting cloud projects, one of those will work… and then build videocoin to live on / by using a distributed cloud.
i was digging into the whole videocoin thing, from my understanding a CDN basically takes video from one location, transcodes it and the maybe stores it in some other location in which it can be streamed from, or which is geographically conducive to improve network utilization.
like say a new movie stored in china was popular in the states, then the CDN would transcode it and store it somewhere in the states, so that the network utilization is much better.
i’m sure it’s much more complex than that, and that my knowledge of videocoin only scratches the surface… but it made sense to me this way.
it just sort of made me wonder if videocoin is actually pretty close to what is being suggested and if that is using tardigrade for it’s network / storage then it might already be an option when hollywood or whatever uses videocoin as their CDN.
Sooner or later there will be more information available on Videocoin.
Here is something that I have found:
In phase 1, which is currently advancing, the VideoCoin Network development team will be building tools to allow for the storage of video files, made capable by Storj. This is the critical first step in building the technical foundation needed to integrate key features via this partnership.
The development work being done includes feature additions in Storj software stack to make the live-streams generated on VideoCoin playable from Storj’s storage. We will be providing python scripts(open source repo) to move the files generated on VideoCoin to Storj. The publishers who are using VideoCoin console will run these scripts as a post-processing operation.
VideoCoin Network’s Innovators Program assembles best-of-breed next-generation video solutions to enable the decentralized future of video services - a growing global need in light of COVID-driven video demand. For media customers of the VideoCoin Network, Tardigrade can provide both long-term, reliable storage for transcoded assets and serve as origin servers for customers’ preferred CDN providers.
When a video stream is sent to the end-user, that video must be transcoded. This is the compute-heavy processing of raw data executed by our VideoCoin Network Workers. Workers do the heavy lifting and offer the customer an efficient and cost-effective way to transcode videos.
And from their website: https://videocoin.network/
Video File Encoding
Ingest and encode your video files and output an HLS stream
Livestream With Ease
Stream live from any RTMP or WebRTC source and output an HLS stream
So it seems that Videocoin targets the transcoding of content and is more geared towards streaming. So this sounds more like “Netflix” type customers aka a specific distribution channel for video content.
3 Likes
it’s certainly one of the more interesting projects i’ve seen in relation to this kinda "big screen / small screen) streaming thing, only thing i have difficult understanding is the limit information about it, even tho it’s like a project thats like 4-6 years old or whatever… or thats the oldest video on youtube about it…
ofc hyper scale infrastructure takes a while to plan, build and test… so maybe that’s why…
can’t help but make me suspicious, ofc maybe there wasn’t any cloud storage that fit the bill for their infrastucture and they sort of just started with a long runway.
planing to go more public when the required technology hit the market, which storj did recently…
very interested in this project even if i think it sounds a bit like a ripoff paying … what was it thousands of $ to even join the project.
but i might just do that when i got some liquidity to towards a gamble ish thing.
Generally, a CDN doesn’t include any compute operations on what’s stored on it but does do shifting or performant hosting in a region that is experiencing a higher load.
I mean it sounds what’s needed is just something like a blockchain version of what could be GoFilex/Cino, where Storj is just the CDN with a wrapping DRM platform around it for rights management and billing purposes.
But also though… from a theater quality production perspective, you wouldn’t want to be transcoding anything, since every time you do there is a possible quality loss or long running compute task to not lose quality because you’re running at the highest settings. Studios actually have “farms” of servers to render out their stupid high quality stuff just down to your 1080p blu-ray without losing tons of quality because it can be so compute intensive.
well we have to contend with that if we don’t beat corona which i kinda doubt, then this might last a decade… you think there will be cinemas in the way we think about them then…
i hardly know anything about CDN’s pretty well versed in basic streaming stuff, i do know that how movies was stored and transmitted in the past will not be like in the future…
i duno what solutions will be the go to… but i’m sure that could be calculated by logical deduction.
or ability to process and store data is exploding, soon i doubt regular movies will be any problem to store, then it will be some sort of 3d universe and the “movie” part would be the event cascades / things taking place in the scenes…
i mean sure movies are cool, but why wouldn’t one make them so one could move around in them freely, when it becomes technically possible… this would also be highly useful for historical “simulation” / recreation, and will lead to many new forms of science / archaeology / historical interpretation.
so yeah movies can be quite big but really… we are already getting to the limit of how accurately we can really see it anyways, i’m sure there will be something like cinema’s, but i think the current form will continue like theater is still alive a well… it’s a fine form of entertainment… but these days it’s getting more and more old school.
2D what millennium is this… lol
i think videocoin uses blockchain for sorting the videos
saw the film made from old nasa apollo films… was amazing, hard to imagine it’s that old… film was a pretty hardcore media back then… lol digital storage of movies have until recently been pretty difficult as we can see with all the crappy digital stored old tv stuff that is found all over.
solidified into the best quality affordable for them at the time, and today it just looks so bad it’s barely worth saving.
Just found and posted some more stuff on Videocoin: Videocoin - what is it and how to use
It has some background information.
1 Like
Here is an interesting piece about the movie industry and their preservation problems. It gives some interesting facts on numbers on storage huge media companies have to deal with and how much they pay.
Also might be interesting to talk to: http://www.film-foundation.org/mission-statement
It seems there is desperate need for long term massive storage capacity.
1 Like
one might need to come up with a full solution… maybe something with streaming… storj seems great for streaming… sure it might be slightly expensive at present for most streamers… but one could also imagine it as a sort of semi cold storage from which it could be retrieved quickly when needed…
like if we say popular xmas movies, they wouldn’t get used all summer and thus would sort of slide to the tail end of storage, which often makes it slow to react… tardigrade has insane bandwidth… so it could be imagined as a secure long term storage… not for distribution for millions of viewers, but for storage inbetween when millions of viewers want to see it and when it’s needed then datacenters or videocoin nodes download it for distribution…
i duno… distributed storage has many advantages and making use of them could show some very interesting new markets that might not have existed before… even if the conceptual ideas are just rehashes of old ones in new scales or implementations.
I think one thing is clear: A crazy amount of data gets created per movie. The linked article says 350 TB including deleted scenes and literally anything that has been recorded:
All told, the digital components of a big-budget feature can total 350 TB
These data the companies involved have to work with. This means collaboration, data sharing and transferring it back and forth around the world to any party/contractor or subcontractor that is involved in the processing, finishing and polishing of the final movie.
Then final digital assets needs to be manufactured. This requires data transfers as well and storage. Multiple versions needs to be created for cinemas, TV or streaming and distributed.
After that most of the data has to be archived and reconsidered from time to time if archiving of all of it is still required or if data can be deleted.
This is now my understanding from the stages of the life circle of those digital assets. And I believe Tardigrade could be useful during some if not all stages of it. In any case the amount of data that potentially gets stored is so crazy high, that it probably would pay off to tailor Tardigrade even more to make it suitable especially for this industry. This could include special pricing for archiving material as this sounds more like cold storage.
2 Likes
It all hinges on Tardigrade being a proposition that is so much better that it makes the change worthwhile.
There already exists an infrastructure. Changing systems takes time, developer hours and is not without risk. The financial gain has to be significant to be worth the effort.
If they’re using S3 buckets then it seems like a no-brainer. If they’re using a bespoke system then it’s unlikely they will change.
so basically @direktorn was right in the beginning of the thread… 
funny how people who claim being in a particular industry actual know what they are talking about hehe
that sure is a big amount of data to work with… tho i don’t see why tardigrade couldn’t support that… only issue would be the whole limitation on their local bandwidth…
we should do a test next year on the max bandwidth tardigrade can handle sustained download…
i mean in theory it’s potentially limitless until we run into the collective bandwidth limit or some kind of internet bandwidth limit.
if nothing else it would be fun to see…
The argument wasn’t about the unedited movie size was, it was what the theater plays and downloads to play. Epically if were talking about a TV show with more then one season. The size can rack up pretty quick.
i duno what type of formats they move around for what and for which reasons…
but yeah sure i guess, and an edited version would ofc take less space… atleast if we are consider if its unedited… but not like the scenes wouldn’t get more complex and sound would be added and also be much more complex…
an edited film might not be 350TB but i’m sure it’s enough to be unrealistic for most to work with.
i mean it’s been 10-20 years since i did anything in video and i do remember it being insane back then when uncompressed…
2006 for blueray 50 gb for a two sided, and that is a home media storage that is 14 years old… and people try to say that it takes up less space today for cinema’s… to be fair … its outrageous
fair i don’t know much about cinema movie storage…
but look at the NASA Apollo 11 recently released in 4K made from only original film…
yeah keep telling me that worse than what was recorded in the 1960’s is being sent to cinemas around the world today… the more you say it, the more i believe it… hehe
it certainly wouldn’t surprise me if movies took 1/3 or 1/7 or so of what they where in their raw format, but i highly doubt they will be smaller than that… most compressed has quality loss… ofc i’m not talking low end movie theater, but ofc a nice one with good stuff…
around here some people seem to think a cinema is a projector and a sheet lol
but yeah 16 years ago a home movie took 50GB
Yeah for sure its not really realistic to work with any video files that are 350TB I have a hard enough time dealing with red files that are around 400gigs and up… If I had to transfer these out side of my network it would be alot of work… And I have a 10gig network and my server has massive transfer speeds to itself. It takes far to long. I also got my 3090s to help with editing these monster files.
1 Like
yeah video editing is like getting into container shipping in a sailboat… or something lol
atleast when one goes from the home user setup to the professional type setup
highly specialized setups.
infiniband is the way to go for nice speeds for networking, memory use to be pretty important for that kind of stuff… ofc it depends a lot what one is doing with the data…
if there is one thing i’ve learned over the years with computing… you can almost never get to much memory lol
wouldn’t mind a couple of 3090’s but i think just one would blow out my psu