Frequently asked questions

Hivenet is a secure, distributed, sustainable cloud, federating and weaving unused capacity of computers. Hivenet offers secure and sustainable cloud storage and compute services powered by its distributed cloud network. Users of Hivenet, aka Hivers, share their computing resources, hard drives, CPUs, GPUs and bandwidth with their peers to create Hivenet's network.

A Hiver is someone who actively participates in the Hivenet community by sharing their computing resources, such as hard drives, CPUs, GPUs, and bandwidth, to help build and maintain Hivenet's network. This term includes both individuals who contribute storage and/or computing capacity, as well as the employees of Hivenet itself.

The Hivenet community is a collective of active Hivers who contribute their unused computing resources to Hivenet. They are the backbone of our distributed cloud network, helping to make Hivenet more robust, secure, and sustainable. The community is not just a page on our website—it's a dynamic group of people who actively participate in and contribute to the growth and development of Hivenet.

We are building the following services:

• Store: secure and sustainable cloud storage.
• Compute: Distributed GPU cloud for AI and high performance computing.
  

• Send (tba): securely share your files without any size limits.

Our cloud storage is already available. We will progressively introduce the rest to the Hivenet community as we deliver our roadmap. Each month we’ll be adding new features. Stay informed on our product updates by following us on our social media channels or Discord (check the footer for all the links).

When you put your files in Hivenet, these files are split into chunks. Each chunk is then compressed, encrypted, and encoded for data reliability before it is sent over to the cloud. The reverse happens when you download your data from Hivenet. Read more

Yes. Hivenet works on any type of computer as long as it runs one of our supported operating systems: MacOS, Windows 10 or 11, and Linux.

Hivenet's performance is not sensitive to the number of nodes that are part of Hivenet. Peer-to-Peer systems, like Hivenet, offer intrinsically better upload and download speeds than traditional client-server systems.

Technically we do not have a hard limit for file size that the platform supports. You can safely upload a single file up to 100 GB and more from desktop (Mac and Windows). While we haven't extensively tested larger files due to storage costs, we believe we can potentially handle files as large as 500 GB.

For mobile uploads, the situation is a little different. Currently, you can safely upload files up to several GBs. However, this requires the app to remain in the foreground without entering sleep mode, which can be challenging for most of you who have sleep mode enabled on your devices.

Compute is now in its beta phase. You can request a demo here.

We are building Hivenet for everyone and expect that individuals and businesses will benefit from our secure storage and compute solutions. With each release of our software we will create richer features which will cater to more specific business requirements.

Yes. You can install Hivenet on multiple machines.

Hivenet uses distributed computing technologies, known as peer-to-peer computing, to enable users' computers to communicate amongst themselves rather than through a central authority, i.e. datacenters of a centralized cloud provider. These peer-to-peer technologies have been in the market for over 30 years. The most known examples of applications using peer-to-peer computing are: Skype, BitTorrent, and Bitcoin.

Hivenet is engineered to minimize how much you need to trust us. We’ve built security and confidentiality into the core of the system. Your encryption passphrase is used to generate your encryption keys and it never leaves your device, and isn’t accessible to anyone, including us. All the data that you store on Hivenet is encrypted, including its metadata, with this encryption key that only you will ever have access to.

Additionally your data is cut into small "chunks", each of which is encrypted again, this time with a different key, before being distributed throughout Hivenet. This means that the possibility of an attacker gaining access to your files and then having the ability to decrypt and assemble each of the chunks to recompose your file is practically impossible.

Hivenet users have two passwords. One for the login, the “account password”, and one for the encryption of the files, which we call the “encryption passphrase”.

The account password is managed by Hivenet, and can be recovered. It is not used for anything else but logging in. This password cannot decrypt or encrypt any files or associated metadata in Hivenet.

The encryption passphrase you create when you set up Hivenet for the first time is only available on your computer, and nowhere else. Hivenet cannot access it, therefore it cannot be recovered if you lose it. If you lose it, you will lose permanent access to all the data stored in Hivenet. Keep it safe, for example in a password manager.

We’re in the process of establishing partnerships to demonstrate that we are true to our word with:

• Security agencies
• Auditing agencies
  
• Pen-testers

We’re also putting in place a security bounty program to ensure that, if you or anyone finds an issue, we will happily collaborate with the finder, and reward them for their contribution.

It may seem counterintuitive, but it’s actually better for performance to store files across many distributed nodes. If stored on a single node, you will retrieve a first block of data, then a second, etc. If these blocks are stored in many nodes, you will retrieve them all at the same time. Performance actually depends on many other parameters (your access bandwidth, the load on your device or the server, how many back-and-forths are required to retrieve your file, etc.). But the peer-to-peer nature of our solution provides intrinsically better performance than centralized systems.

Hivenet's system is resilient to nodes appearing and disappearing from the network. Imagine your file broken into 100 pieces, like a puzzle. Hivenet builds an additional 100 special pieces that it distributes with the original pieces. As long as Hive can retrieve 100 out of the 200 pieces, it will be able to reconstruct the puzzle.

While some devices will be on or off, other devices such as NAS or mobiles are mostly always on. Hivenet understands the behavior of the devices and assigns a class of availability to each device. This enables a proper placement strategy to distribute these 200 pieces to maximize availability.

Hivenet first guarantees that the data in each puzzle piece is not corrupted by storing signatures along with the data. Corrupted data is then removed from the network. Second, Hivenet special nodes monitor these puzzle pieces, and reconstruct new ones when not enough pieces are available. Over time, you will always be able to reconstruct the puzzle.

There are approximately 20 million servers in data centers around the world. These consume today approximately 40 TWh representing 2% of overall electricity demand. However, due to the growth of the Internet - storage and compute - this is expected to grow to up to 20% of overall electricity demand in the coming decade. This is soon going to outgrow the energy consumption of the airline industry.

A significant portion of this energy is not directly used for useful compute and storage:

• 43% of the energy is consumed for cooling and by idle servers.
• 14% of the energy is consumed by the storage disks and network.
  

Hivenet is designed to save energy on a number of elements:

• No dedicated cooling is required on user devices.
• No additional servers are needed to run programs.

People will turn off their devices (or let them go to sleep mode) when they get idle. We will get power savings, at the expense of adding more data redundancy in the network.

Lastly, Hivenet will take into account geographic proximity when allocating node storage to ensure that data avoids traveling the entire planet to get to its destination.

We don’t expect people to keep their computers powered. Hivenet is designed to be resilient to intermittent nodes at the cost of adding data redundancy in the network to compensate for it. Hivers should actually keep using their computers as they normally use to: their power consumption will then be linked to actual real activity.

Devices that are always-on (e.g., servers, etc.) will be classified as such and used in priority to avoid unnecessary data redundancy.

In short, both.

Even though these two terms are often used interchangeably, they actually refer to two different concepts:

• Centralization and decentralization refer to control: in a centralized system, control is exerted by just one entity, which is not the case in a decentralized system, where control is exerted among several independent entities.
• Centralized and distributed refers to differences of location. In distributed systems, parts of the system exist in several locations.
  

With the advent of distributed ledgers such as blockchain, decentralization has been highlighted as the main innovation and differentiation. You will often read that “distributed ledgers eliminate the need for central authorities”. Indeed, but this has also created some confusion. Some “permissioned or private blockchains”, used for businesses, may be only partially decentralized, because they rely for some aspects on a central authority (like the ability to join the closed network).

Hivenet is a distributed peer-to-peer cloud, as it is hosted on everyone’s computer. It also relies on a decentralized data exchange layer: a Hiver will always be able to retrieve her/his data directly from the peers, without interacting with any central authority like Hivenet. However, like for “permissioned blockchains”, some of our advanced services will rely on Hivenet's control layer.

Yes, Hivenet is a web3 initiative, as our product is built on a distributed computing framework, but unlike many other similar initiatives, Hivenet doesn’t use any of the blockchain technologies.

Web3 is defined as a decentralized system with data exchanged peer-to-peer, which relies on devices at the edge (mobile phones, appliances, sensors, laptops), which is trustless and permissionless, and semantic-based. By this definition, Hivenet fits into web3.

No, Hivenet does not have its own crypto coin.

GDPR is a legal framework setting guidelines for the processing of personal information. On Hivenet, all personal data stays with our users; the only personal information our Hivenet collects are the person’s name and email address. All other information on Hivenet is encrypted and cannot be linked back to a user, or even a user identifier. Beyond that, we provide the capability for our users to modify this information and delete their account and the corresponding data.