tfgrid4 update farming

This commit is contained in:
Mik-TF 2024-05-08 15:04:24 -04:00
parent 9e7d698aee
commit 74038afb0d
4 changed files with 63 additions and 59 deletions

View File

@ -1,13 +1,13 @@
![](img/bl_7840c.png)
## ThreeFold Farming = **Become a Cloud Service Provider**
Farming means a person offers compute, storage and network capacity to the network
![](img/bl_7840c.png)
Farming means a person offers compute, storage and network capacity to the network.
- Farmers buy a 3node or mnode (500-2000 USD)
- A farmer connects the node(s) to internet and get rewards for providing capacity
- A farmer earns rewards.
- A farmer is helping his/her community to get access to sovereign Internet & Cloud applications on an unbreakabale co-owned Internet/Cloud.
- A farmer earns rewards
- A farmer is helping their community to get access to sovereign Internet & Cloud applications on an unbreakabale co-owned Internet and Cloud
### Income for the Farmers
@ -21,71 +21,79 @@ The recommended pricing is:
- $0.005-4 for Compute/GPU Capacity (gpu, AI, VM, ...) (per hour)
- $0.01-0.2 for 1000 messages over LoraWAN (long range wifi)
> 80% of these fees goes to farmer, 20% to the *TFCOOP, exception for a twin which is 50-50% split.
> 80% of these fees goes to farmer, 20% to the *TFCOOP, exception for a digital self (twin) which is 50-50% split.
### Farming Boosters
Our aim is to have millions of nodes in the field to evingly create the biggest network of Cloud & Internet Capacity in the world, depending the amount of nodes in a TFGrid Cell.
Our aim is to have millions of nodes in the field to eventually create the biggest network of Cloud & Internet Capacity in the world, displaying the amount of nodes in a TFGrid Cell.
Cell's get colors in line to need of service:
- green (enough nodes, no rewards)
- yellow (can use some extra nodes)
- red (let's do it, this region needs help)
- Green
- There is enough nodes
- No rewards
- Yellow
- There is a need for some nodes
- Rewards
- Red
- There is a big need for nodes
- This region needs help
- Highest rewards
For red cells the reward is highest.
### How to Become a Farmer
- Buy a *TFNODE (Compute/Storage node) or *TFROUTER (Mycelium Network Node).
- Build your own TFNode
- Our software makes it easy to get started.
### How to become a farmer
### Rewards for Being a Farmer
- buy a *TFNODE (Compute/Storage node) or *TFROUTER (Mycelium Network Node).
- Assemble your own TFNode, our software makes it easy to get started.
### Rewards for being a Farmer
Typically a Farmer can earn the investment of their node back 2 to 10 times over 5 years if succesful.
Typically a farmer can earn the investment of their node back 2 to 10 times over 5 years if succesful.
> For a [simulation based on twins with storage and bandwidth check here](https://docs.google.com/spreadsheets/d/1D6Q3Yav_SS356zAILwN4SrmqwYfLibT3E6oWrVPU4c8/edit#gid=170998100)
### Farming Pools
Some farmers will chose not to manage, market or operate their 3node, in that case they can join a farming pool.
Some farmers will chose to not manage, market or operate their 3node. In that case, they can join a farming pool.
A Farming pool takes care of
A Farming pool takes care of:
- marketing the capacity
- giving support to the userbase (developers)
- managing the best possible pricing
- monitor all nodes and optimize uptime
- monitor available bandwidth and make optimizations where possible.
- Marketing the capacity
- Giving support to the userbase (developers)
- Managing the best possible pricing
- Monitor all nodes and optimize uptime
- Monitor available bandwidth and make optimizations where possible.
If a farmer uses a farming pool he/she will have to give a part of their reward to the farming pool, each farming pool can decide how much that needs to be.
If a farmer uses a farming pool, they will have to give a part of their reward to the farming pool, each farming pool can decide how much that needs to be.
### Advanced: Compute Box
The more advanced farmers can chose to define compute boxes in their node. A compute box is an amount of CPU, MEM, GPU, SSD as specified by farmer. When users select a node to deploy a VM on they can use one or more of these compute boxes.
Example a node with 64 GB or mem and 2 TB of SSD and 24 virtual cores.
Example a node with 64 GB of memory, 2 TB of SSD and 24 virtual cores.
- 15 compute boxes each:
- 4 GB of memory (60 GB total)
- 120 GB of SSD capacity
- 6.4 logical CPU core (oversubscription of 4, which means user can max use 4x CPU capacity if system allows)
- when a user choses the full machine, then he/she will have reserved all compute boxes capacity which means the machine is now dedicated reserved for the user, the hoster specifies the discount for this typically 50%. On a dedicated machine the user has full access to the GPU.
- min 1GB always needs to be left as buffer for memory and 10% of SSD capacity
- when a user choses the full machine, then they will have reserved all compute boxes capacity which means the machine is now dedicated/reserved for the user, the hoster specifies the discount for this (typically 50%). On a dedicated machine, the user has full access to the GPU.
- As a buffer, a minimum of 1GB of memory and 10% of SSD capacity are needed
How does it work:
- When a developer wants to deploy a virtual machine, he/she needs to define the required capacity and will be able to make a choice based on these compute boxes, each compute box has different base specs (mem, ssd, gpu, cpu) and reputation, the developer can now make a selection of how many of these compute boxes need to be given to the virtual machine. This defines the monthly price which will have to be paid as well as capacity available to the VM.
- When a developer wants to deploy a virtual machine
- They need to define the required capacity and will be able to make a choice based on these compute boxes, each compute box has different base specs (mem, ssd, gpu, cpu) and reputation
- The developer can now make a selection of how many of these compute boxes need to be given to the virtual machine.
- This defines the monthly price which will have to be paid as well as capacity available to the VM.
## Remarks
### Network Rewards requires Unblocked Incoming Internet.
### Network Rewards Requires Unblocked Incoming Internet
"Incoming internet traffic should be enabled to earn rewards from network bandwidth.
Incoming internet traffic should be enabled to earn rewards from network bandwidth.
If a farmer lacks this capability, their TFNode will connect to a neighboring TFNode with good internet connectivity and public incoming access. In this case, the neighbor farmer will earn money from the internet traffic, while the farmer will continue to earn rewards for compute and storage."
If a farmer lacks this capability, their TFNode will connect to a neighboring TFNode with good internet connectivity and public incoming access. In this case, the neighbor farmer will earn money from the internet traffic, while the farmer will continue to earn rewards for compute and storage.
All traffic is end2end encrypted over the Mycelium network = our P2P Network which lives on top of current Internet.
@ -93,7 +101,7 @@ All traffic is end2end encrypted over the Mycelium network = our P2P Network whi
- IPv6 public ip address available
- IPv4 public ip address available (harder)
- Portforwarding or hole punching support for UDP.
- Portforwarding or hole punching support for UDP
Mycelium will try a lot of tricks to allow traffic to go to your network.

View File

@ -1,8 +1,7 @@
# network map
# Network Map
ThreeFold aims to have a map to visualize all of the ThreeFold nodes, routers and devices, from farmers and partners.
As an example, Helium offers the following map:
![](img/networkmap.png)
Above network map is from helium see https://explorer.helium.com/ but ThreeFold aims to have a similar map to visualize all threefold nodes and routers and also devices from our partners like Helium.

View File

@ -1,7 +1,7 @@
# Available TFNodes
> TODO: explain we can create nodes ourselves
<!-- TODO: explain we can create nodes ourselves -->
### TFNode Silver
@ -11,7 +11,7 @@
- 512 GB SSD
- min 4 logical CPU cores
Will cost less than 500 USD.
It will cost less than 500 USD.
### TFNode Gold
@ -21,8 +21,7 @@ Will cost less than 500 USD.
- 1000 GB SSD
- min 8 logical CPU cores
Will cost less than 1000 USD.
It will cost less than 1000 USD.
### TFNode Platinum
@ -33,19 +32,21 @@ Will cost less than 1000 USD.
- 4000 GB HDD
- min 16 logical CPU cores
Will cost less than 1800 USD.
It will cost less than 1800 USD.
### TFNode Datacenter
- put your own TFNode in a datacenter of choice, make sure you have great and cost effective reliable connectivity to internet.
- The Farmer buys their own hardware, ThreeFold supports the farmer to get up and running.
- Put your own TFNode in a datacenter of choice
- Make sure you have great and cost effective reliable connectivity to internet
- The Farmer buys their own hardware, ThreeFold supports the farmer to get up and running
### TFNode Dedicated Rental
- rent a node in a dacenter, make sure you have good capacity in that datacenter. Companies like OVH and Hetzner come to mind as reliable and cost effective.
- Rent a node in a dacenter
- Make sure you have good capacity in that datacenter
- Companies like OVH and Hetzner come to mind as reliable and cost effective
## Remarks
We are still working on the exact logistics, the look and feel can be different, but minimal resources will be the same.
We are still working on the exact logistics, the look and feel can be different, but the minimal resources will be the same.

View File

@ -1,7 +1,7 @@
## TFRouters
The TFRouters allow everyone to connect the users of our network with the *TFNODE
The TFRouters allow everyone to connect the users of our network with the *TFNODE.
### TFRouter Indoor
@ -15,7 +15,7 @@ More than 1 node can be installed in a larger office, hotel, home, etc., and the
Each node runs our *ZNET Operating System, allowing for native integration into the TFGrid network.
[Zero-OS](def:zos) Nodes are seamlessly integrated with the Zero-Net nodes by design, requiring minimal configuration. They are part of our Autonomous Zero-OS Stack.
Zero-OS Nodes are seamlessly integrated with the Zero-Net nodes by design, requiring minimal configuration. They are part of our Autonomous Zero-OS Stack.
![](tfrouter_physical.png)
@ -45,14 +45,10 @@ Currently, we rely on fiber or existing internet connections to link between loc
![](img/TFRouter_lr.png)
> We are looking for volunteers who will be rewarded to do reasearch and experiment with chosen equipment to test performance in real life situations, let us know if interest to help.
> We are looking for volunteers who will be rewarded to do reasearch and experiment with chosen equipment to test performance in real life situations, let us know if you want to help.
## Lo-RAN
![](img/loran.png)
!!task.add story:'TFRouter_selection' title:'describe loran on TFRouter'
<!-- !!task.add story:'TFRouter_selection' title:'describe loran on TFRouter' -->