- How is Arken funded?
- What is the Core Arken Cluster?
- Does my data qualify for the Core Cluster?
- How is Arken different from Dropbox, Google Drive, etc...?
- How is Arken different from Amazon S3?
- How is Arken different from IPFS Cluster?
- Can I easily stop contributing space to Arken?
At the moment Arken is fully funded by myself (@alecbcs). Excluding the cost of nodes provided by the community. I currently pay $5.70/month to run the two bootstrap nodes (apollo & artemis), the statistics backend (arkstat), and files.arken.io.
However, we are looking into how we might setup an Arken entity to accept donations and ensure the long lifespan of the project.
If anyone has experience with or can assist with non-profits, fiscal hosts, or building an organization feel free to reach out to us at team@arken.io.
The Core Arken Cluster is the official network hosted by the Arken community dedicated to preserving scientific and open source data. The central manifest for the Core cluster is hosted at github.com/arken/core-manifest.
You can download data from the Core Cluster either by using Ark: the Arken Command-line Client or by going to files.arken.io.
In the future other groups may decide to create their own clusters dedicated to specific areas of research.
Probably! Here are a couple of requirements for scientific & open source data contained within the Core Cluster.
-
All data within the Core Cluster is publicly available. This means that we don't accept encrypted datasets and you should be comfortable with anyone being able to inspect your data. The Core Cluster is made of up contributions from the Arken community and it wouldn't be fair to accept submissions that could not be audited by anyone in the community.
-
Your data must be open source, or you (as the copyright holder) must give permission to Arken legally to store and distribute the data on your behalf.
Uploading data to Arken is fundamentally much more like submitting a paper to a research institution than it is like uploading a file to your other cloud storage solutions. Unlike Dropbox, Arken isn't going to store your baby photos or lecture slides. Arken also won't automatically sync a file between the cluster and your laptop. But if you've got supplemental data for a research paper, or an open source publication, Arken will replicate it globally and make it available to any one in the world.
Object Storage (also often called S3 compatible storage) makes storing and manipulating data incredibly easy on anything from 1GB to 100TB. By contrast data within Arken is immutable, meaning that once it is uploaded to the archive it cannot be changed by storage nodes and any changes to the files will require them to be resubmitted to the manifest repository.
When writing Arken we wanted to build a fully auditable system. That meant the data in the archive as well as the code. We didn't want users contributing their extra hard drive space to have any concerns that we might be put other hidden data on their systems. That's why instances of Arken don't pull any information about the Archive from our own systems. Everything an Arken node needs is in the manifest repository which can be audited by anyone at any time. The only way to add data to can Arken cluster is by updating the manifest. By contrast IPFS Cluster is built for enterprise systems where users with access can add files to the cluster through the API without modifying some publicly available manifest.
Absolutely! You are under no obligation to continue hosting data if your circumstances change. Arken continuously re-balances data across all of the nodes available so other systems can automatically pick up from where your node leaves off.