Token Lock and Farmers Building at Scale

I’m currently in my garage and can run 40 servers. I have a commercial building with infrastructure ready to go to stand up 6 racks(120 servers) total within the next 6 months.

The concern is operating costs and token locking. It will take 20 running servers just to cover rent, electrical, and fiber at scale. If the tokens are fully locked, I can’t cover operating expenses or get my ROI.

Is there any decision yet on whether and/or when tokens will be fully locked for 24 months? Is there a consideration for people like me building at scale? Again, I’m ready to immediately stand up another rack and move into the larger space, but I can’t justify the investment if the tokens are not at least partially released. Need to know ASAP.

5 Likes

I think this is a very good and important question. Indeed it should be addressed soon as it implies day to day farming realities.

Is the DAO ready for such a question? Can we pass a vote, with different options on the following subject?


(How) should TF proceed to unlock partial TFT rewards to cover for Farming Running Fees


Can it be set on a model like vesting was done?
Can it be done with a basic percentage available per month?

Open discussion.

I’d say it’s more a : yes it should - but how exactly - situation.

3 Likes

Hi @colossus. This is indeed a good time to think expanding. When you start / expand farming now you will get the token listing value set at 8 cents. This is a very low price to get a 60 month fix on. Having said that I do get your concern, expanding costs money and that additional OPEX has to be covered somehow.

We’re in a very exciting phase of the where it is a chicken and egg problem. The TF Grid is beta but stable enough to run simple workloads. We get tremendous responses from the like of owncloud and presearch because it simplifies setting up node (single site, single instance) for their communities a lot. Singe click deployment almost on a servers / region of your choice. For these it doesn not matter (besides having a dedicate IPv4) whether they run from the home or a mode professional data-center.

For larger installations, read enterprise workloads, archive, backup, multi-site installations, we ge a lot of interest from Enterprises that see the benefit of breaking out of the walled gardens of market leading clouds and to be able to claim true data sovereignty and governance. For this we need more traditional datacenter based installations (which is what I believe you are building).

The chicken and egg is that we need a sizable grid to be taken serious but this creates farmers that hold the key to the project with regards to token value (eg. speculation / dumping or not). Vesting has been a mechanism to “damp” the speculative aspects of the digital currency markets and provide a stable, appreciating token based on real utility, a utility that serves a for ever growing need om planet earth.

The majority of the community we currently have are all on board with making this a great project and therefore will not dump tokens and damage the project, so the question is do we need vesting as a protection mechanism? It certainly does not help you to come on board and helping solving the sizing issue.

So what to do? I think a workable solution would be to have a percentage of the minted tokens lockup (automatic vesting) and the remaining not. This would create relief for operational costs if you want to cover that with a token amount. What do you think?

5 Likes

I understand that maybe what needs to be done my node is in Germany, tax technically it will be a horror as I cannot move the coins in a other Wallet, I then have to pay taxes differently and may also make losses

I think TFT will lose a lot of nodes from Germany

You can change where your tokens go if moving them from the wallet is a problem.

Hi, @weynandkuijpers. I think you are spot on about the entire chicken and egg scenario. There is certainly a balance needed to protect the stability and growth of the project and token, while also providing OPEX relief for sizeable builds like the one I’m proposing.

I think everyone involved with this project is really onto something spectacular here. My goal is to align with the project’s intent as much as possible by not centralizing too much of the network, but at the same time also providing a solid and reliable chunk for larger workloads.

I believe I’m in the sweet spot of accomplishing this. The perfect storm of opportunity came by discovery of this project along with the simultaneous acquisition of enterprise backbone gear(Cisco Nexus 10Gb core switching and FEX in best practice dual home config protected by Vertiv Liebert APS) and a building wired for 1,200 amps of service. The last few pieces to the puzzle are the acquisition and deployment of 80 more servers, a backup generator, and a commitment to 36 months of dedicated fiber service.

I’m willing to invest a large sum to make this a reality. With a plan to put the bulk of initial profit toward, and meet ROI within 6 months. After my ROI is met, I would be perfectly content with vesting the bulk of tokens and only withdrawing to cover OPEX and upgrades. I’ve calculated that after the ROI is met, and the next 24 months pass, TF will(presumably) be in a very strong position and it will be time for an equipment overhaul on my end. I’m in for the long haul but have to ROI by the end of 2022 to be in a comfortable position. What happens with token locking will be the deciding factor between 40 servers in the garage, or 100 - 120 servers in a best practice scenario.

Lastly, Thank YOU and everyone on the TF team for your vision and execution of something truly next level. I’m honored to take part and look forward to what the future brings.

6 Likes

I love your enthusiasm for the project and for building out your farm. Glad to have you onboard. Hopefully we can find a way to also park a block of IPv4s with all that infrastructure :slight_smile:

Popping in here mostly to say that I agree we need to consider operating costs, including tax liabilities when locking up tokens. This is especially true for farmers operating at large scale. We are imagining three possible tiers for farmers:

  1. DIY
  2. Small certified
  3. Professional certified (redundant power and network, dedicated fiber, etc.)

Farmers in this professional category have substantially higher operating costs than farmers at home or in non dedicated commercial spaces like offices. We can start with the typical cost of a colocated rack versus the farming output of typical rack style node configurations to get a ballpark. My gut off the cuff response is that up to 50% unlocked makes sense. Knowing what percentage you’d need unlocked to meet your anticipated OPEX and also hit your ROI goals would be helpful here too, @colossus.

This may also be a good time to address a dynamic within the currently proposed lock system, which is for 24 months or until a 30% capacity utilization threshold is reached. While the specifics of the latter requirement haven’t been hashed out, it suggests the possibility that farmers may buy up their own capacity to unlock their tokens. Unlocking a certain percentage of tokens for all farmers up front could make this unprofitable, thus actually enhance the model in a way.

3 Likes

Thanks, @scott! Likewise with your enthusiasm and support. I’ll definitely be assigning public IPs to all these hosts.

The Pro Certified category idea sounds like a perfect addition for builds like mine. 50% unlocked sounds like a good balance. It does bump my ROI timeline from 6 months to 12 months which is okay. The only drawback is it will slow my expansion. I had planned to achieve 6 racks(120 servers) by the end of 2022. I don’t know that I could put that much CAPEX out in that period of time at 50% lock.

I’m okay with stretching buildout into 2023, but what will that look like for new nodes coming online a year from now? If I’m understanding things correctly, the earlier you get a node registered, the more its worth? If I have two identical spec nodes, one registered today, one registered a year from now, the one registered a year from now won’t earn as much, right?

Jumping back to the percentage required for OPEX, that’s still a bit tricky considering the nodes are just idling right now. I know how much power they’ll consume at full load and can plan for that, but the bandwidth requirement is still a large unknown. I’m guesstimating that for 120 servers, 10Gb core switching and routing should suffice, but what is my ingress/egress demand going to be? 1Gb dedicated fiber is $1,399/month with only a block of 5 statics included. I need a block of 120 statics plus possibly up to… 5Gb? 10Gb of service? :man_shrugging:

The thing to remember is that the rewards per 3node are set by taking the token price in considerations. The 3nodes rewards are set in USD (see here):
image.

When you register your 3nodes in the blockchain the (current) token price is recorded as well, and set for 60 months. This means you will have a fixed reward for the coming 60 months. The token prices is expected to appreciate going forward, so registering 3nodes at a later point in time will likely give less reward token.

At some point in time the proof of capacity algorithm will include “proof of sufficient” networking capacity. This is not the case today, and this will be proposed as a GEP (grid enhancement proposal) on this forum in the (near) future. For now here’s some thoughts that have been discussed recently.

1 Like