Remote sensing clients

[Garfield]

Has the STORJ sales team reached out to clients with satellites (the things in space)? I know that the remote sensing industry has many PB’s of data.

[Knowledge]

Thanks for the suggestion! I’ll let the team know your thoughts.

[Alexey]

Yes, we have had contacts with NASA several years ago. However, I do not know the current state.

[Garfield]

Great to hear! I was also thinking about other GIS related firms/goverment. Examples could be:

• Kadaster (https://www.kadaster.nl/)
• Cyclomedia (https://www.cyclomedia.com/nl)
• PDOK (https://www.pdok.nl/)
• Nationaal Georegister (Nationaal georegister)
• Esri with their software ArcGIS
• Planet Labs (https://www.planet.com/)
• ICEYE (https://www.iceye.com/)
• Satellogic (https://satellogic.com/)
• Maxar Technologies (https://www.maxar.com/)
• Airbus

[jammerdan]

@Alexey

Yes, we have had contacts with NASA several years ago.

This?

NASA needs 250PB storage by 2025 and cannot afford AWS

Could be great tardigrade opportunity: https://www.theregister.co.uk/2020/03/19/nasa_cloud_data_migration_mess/

Maybe worth to revive?

[Alexey]

I’m afraid it may end in SOC2 facilities (Storj Select)…

[Garfield]

The team probably already knows this, but of SOC2 keeps popping up as hurdle in the sales, that’s the next priority for STORJ. Or is getting SOC2 not possible?

[xsys]

CERN - https://home.cern/news/news/computing/exabyte-disk-storage-cern
ESA - https://www.esa.int/About_Us/esait/Big_Data_for_ESA
This kind of customer may want to even run their own satelite…

[bre]

Storj Select offers SOC2

Compliant cloud storage

Introducing Storj Select: Revolutionizing Cloud Storage

[Garfield]

Yes, but i meant for the public network

[Alexey]

We are working on it.

[jammerdan]

@Alexey

I’m afraid it may end in SOC2 facilities (Storj Select)…

Better there than on AWS?

[jammerdan]

I had made a similar suggestion in the past:

Satellite as a service

Therefore, Blackburn is always on the lookout for another method to move petabyte-scale data. I am sure they are aware of cloud offerings like Wasabi and others which are cheaper than AWS. So probably that’s still too expensive and not a good fit for their use cases. It was even said that the final data lies on Googles cloud. But what if the offer would be that they could build their own data center by running their own satellite? I am sure there are hundreds of thousands of astronomy ent…

Storj DCS for scientific and research communities

The new great review of Storj DCS by Prof. Portelli made it clear that Storj DCS is a great storage for large datasets that need to be shared around the world by research and scientific communities. The big question is, how Storj can come into contact with those and offer Storj DCS as storage to them. I came across an European initiative that aims to bring together providers of all kinds with users from research, science and universities to form an open cloud for sharing and accessing research…

I am pretty much convinced that there are millions of astronomy enthusiasts who would be happy to donate some disk space to a project that provides cheap and reliable data for that field of science and hobby. Maybe even for free. Storj could setup the satellite and get paid for running the satellite, while the researches can have access to less expensive space for all of their data. Powered by enthusiasts like we have seen with projects like Seti or FaH…

[Ottetal]

Who cares? It would be a massive client for StorJ which both make the future outlook of StorJ bigger, it increases the amount of engineers StorJ can take on, and it’s a huge (!) green flag for other prospect customers

[Roxor]

If any popular project wants to host files in Storj-S3 using “free space” donated from their community… they can just start using the vanilla public network like any other customer. Then publish their Storj account ETH-deposit-address for anyone in their community to use in their own node. Anyone using that as their node payout address will be helping pay for the projects use (and the project gets the 10% bonus from payment-by-token).

• The project pays nothing out-of-pocket (and would likely maintain a wallet credit)
• The community donates their space (and it enables the project to share all the public files it can afford)
• Nobody has to mess with a custom satellite or special-subset-of-SNOs… so everyone gets the performance and durability of the public network

Win, win, win? Don’t overcomplicate it…

[heunland]

Please note that deposits would need to be in form of STORJ, we do not accept ETH as form of payment.

[Alexey]

Actually I believe it’s a great idea, especially if to use a zkSync Era L2 network for STORJ tokens.

[jammerdan]

And what happens to the number of nodes? Will we then have 100k nodes in no time?

[Ottetal]

That’s a wonderful way to solve the problem!

[wthorp]

I previously worked at Airbus Defense & Space’s US office. Ten years ago, they had 6 PB on local spinning disk, but that operation was moved to GCP. Similarly, Digital Globe - now part of Maxar - moved operations to AWS and was the one of the first Snowball clients. I’m not sure about ESRI, but Azure is a likely target given ESRI’s focus on county-level government clients typically running Windows. Larger US imagery providers such as NASA have some of their data hosted via Amazon’s Open Data Sponsorship Program. These petabyte scale imagery providers likely have entrenched cloud solutions with well negotiated pricing.

Diving deeper, the traditional API used for satellite imagery manipulation is called GDAL. GDAL includes a virtual filesystem with an S3 driver, so it should work well with Storj. It’s been a long time since I’ve looked at the code, but GDAL used CPU for image manipulation, not GPUs. It tends to be paired with other CPU-centric frameworks such as Numpy for color correction across imagery sources, etc… Thus, the historic lack of compute offerings at Storj limited some interest. With the acquisition of Valdi, this is changing. While we’re unlikely to offer on-demand CPU services in the near future, we happily negotiate longer term CPU leases for larger customers.

All this said, Storj’s is a great platform for geospatial data. We have customers working in the remote sensing / GIS fields; they are happy and growing. Like all markets, we will bring on smaller, innovative customers at first and gain the trust of the broader market over time. If you are working in the GIS space, please reach out! I’m eager to see GIS use cases succeed at Storj!!