If you've ever used a cloud storage solution like Amazon S3 and Azure Blob Storage or a managed cloud database like Amazon RDS and MongoDB Atlas, you've surely noticed that costs can quickly skyrocket.
The truth is that their pricing model is quite complicated. You have to consider their offer as a vertical solution, meaning they're not just providing storage space but also taking care of the entire stack. So while you don't need to worry about servers, network connectivity, and maintenance costs, you actually pay for more than just storage space.
In addition, it is tough to compare offers between cloud providers with all the different parameters you have to consider (regions, tiers, data transfers, backup, analytics, etc.). It is even more challenging to forecast how much you will spend because you usually can't control the incoming flow of data, especially if it comes from the simple fact that users use an application of yours.
Undoubtedly, when dealing with personal data, you add another layer of difficulties because of the sensitive nature of the information that can be extracted from it.
That's why regulations like GDPR exist: to protect users and help enterprises implement the right processes for processing personal data. But full compliance is hard to achieve and time-consuming, which induces further expenses. Indeed, doing it well has many engineering and organizational implications that eventually add to the cost of just storing personal data.
But usually, user data is stored in cloud databases which is a specific case of cloud storage that is even more expensive and complicated. The main reason is that with this solution, you also have to pay for computing power (i.e., CPU and RAM) in addition to storage space.
It also means that you add another party or subcontractor in the data's chain of custody, which can be both a financial and an operational burden, depending on your requirements. For example, if you have compliance requirements for GDPR or HIPAA, then managed cloud databases definitely add an overhead.
The idea behind crowd storage is that every user should own and control their data, i.e., how you share it and with whom. It also comes from the realization that much data is duplicated, like our home address, phone numbers, etc., resulting from the incapacity of applications and services to communicate with each other. It means, for example, that when you move to a new home, you must change your address on every service that needs it.
That's why crowd storage is based on the principle of decentralization from a business's traditional point of view but is more like a "re-centralization" from a user's perspective. In practice, for a company, it means that instead of storing their users' data in their infrastructure, they build applications that interact with storage owned and controlled by each user.
Both companies and users benefit from this paradigm because, on the one hand, users have complete control over their data which means that companies don't have to worry anymore about GDPR, and on the other hand, the actual cost of the storage is shared between them, which overall seems fairer for everyone. It also makes a system more resilient to outages and malicious activities.
Myne has a very pragmatic and rational approach: we offer a purposely built (with Rust) federated graph database with a zero-knowledge architecture, which means that only users (not even Myne admins) can access their encrypted data.
Our hybrid cloud database also features real-time queries, data streaming, federated data sharing, and query hooks and works like a headless Content Management System (CMS).
Finally, we are a mission-driven company focused on making the web more open by enforcing data interoperability, which is also built into our database engine.