1st Flora Swarm proposal 🌺

Flora Swarm

Proposal Rationale

Flora has been around for quite a while now but has never been established as other swarms. After our exponential growth, we are in the necessary to enforce and improve all of our infrastructure and services to keep up with demand.

For those who don’t know about Flora, this swarm works deploying and maintaining secondary services, such as validator nodes, notification services, graph node, network nodes, etc.

Servers are something that can’t be decentralized just yet since some of the core services rely on private keys and all of this has to be handled encrypted in an environment with relatively high-security measures (cloud providers). Thus, some trust needs to be delegated to community members (in this case, Flora members) to maintain all of it.

The current members of the Swarm are @crisog, @anisoptera, @lkngtn, @gabi & @onbjerg.

All services are running using Google Cloud (even before the new user wave in September), so we are probably sticking with it unless there is a major reason not to do so.

Servers also imply service costs (paid with a credit card), maintenance costs, etc.

Service costs: they have been covered by Luke for several months. This should no longer come out of the pocket of any member, rather from the funds of the swarm.

Maintenance: this is not covered by anyone, since Flora has been maintaining all of this without any return.

Proposal Funding Information

This swarm has never requested funds in over 6 months of operating. At this time, this is no longer sustainable and we need to create a proposal.

Ideally, this proposal also includes reimbursement for past unpaid work & payments from credit cards out of pocket.

1. Google Cloud billing reimbursement (luke’s cc).

Reimbursement: 
$900 spent since May 3th last year.

Projection: 
$250-350/mo.

Services running:
- Zonal Kubernetes Cluster (4 nodes, 8vCPUS/16GB RAM).
- Two BrightID Nodes (2 vCPU + 4GB RAM each, 120GB)
- Github Actions Runner (for Pollen, 2vCPUS/10GB RAM)

3. Postmark billing reimbursement (luke’s cc).

Reimbursement: 
$30 for three months.

Projection: 
$10-50/mo for 10-50k emails a month.

This is being used at Celeste/Forums.

5. Vercel billing reimbursement (luke’s cc).

Reimbursement:
$60 for three months.

Projection: 
$20/month for all of our static site deployments.

6. Ani’s work (setup & maintain xdai validator node)

Ani's Infra (currently running xDai Validator, but we can use it for more stuff)
$200/mo.

3 months since xDai validator is running, 16hrs (8hr/mo).
Rate: $50/hr
Reimbursement: $1400

7. CrisOG work (setup & k8s cluster/deployments & omnibridge)

Our services/cluster have been running for quite a long time without having any funding,
I tried to estimate a reasonable amount of hours spent doing that work.

A while back I configured/deployed a few token bridge contracts, updated Omnibridge UI so we could bridge $HNY to mainnet, as well as verifying our token on Etherscan and this was supposed to be included within Flora despite not being directly related to infrastructure.

40hrs incidents/maintenance over several months (maybe 7?).
$200.00 flat fee for HNY on Mainnet task.
Rate: $50/hr
Reimbursement: $2200

What’s next?

One of the reasons for this proposal is that there are some big plans regarding infrastructure and we didn’t want to move forward without showing up to the community.

  • Graph node.
  • xDai node (rpc endpoints)
  • Monitoring dashboards for most of the services.
  • Omnibridge & xDai Bridge validator node.
  • Migrate a BrightID Node + Github action runners to Ani’s servers.

However, all of this needs to be always working and if they fail, the community expects it to be fixed in a reasonable amount of time. This availability can’t be measured by hours of work.

$100/week - Ani (Own Infra, xDai Validator Node)
$200/week - CrisOG (K8s Cluster/GCP, rest of services mentioned above)

for incidents availability/monitoring health checks.

All these plans are also subject to security measures to set all of this infrastructure to ensure the safety of our services since some of them involve key management.

10 HNY ~ 7,500 USD for planning & implementation 
This amount will be allocated in hours of work.

Totals Reimbursements (billing/past work) - $3,600 ~ 4.8 HNY

Projection of Flora operation costs (8 weeks) - $1,780 billing & flat rates + $5,000 work ~ 9 HNY per 4 weeks, 18 HNY in total.

HNY Price at the moment: $750

Amount of HNY requested: 32.8 HNY

Vote for this proposal: Honey Pot Proposal

We decided to go with a bi-monthly proposal since a monthly one is really overwhelming, and 3 is most likely a lot given $HNY volatility.

We have a Flora multisig to hold the funds and distribute accordingly.
https://xdai.gnosis-safe.io/app/#/safes/0x91d8116fA60516Cf25E258Ef14dEaAcAf7a74127

We will also create a Kanban dashboard/Github project for visibility.

16 Likes

3 Likes

It is amazing to learn there is so may selfless community members that are working in behind the scenes and contributing effort and funds to keep this ecosystem going. Good on you guys. I wish I had more HNY to be voting but hopefully the whale wallets have enough to approve this proposal! I think there is a few competing proposals and with Celeste released, the existing HNY may be getting stretched to vote in all these really useful proposals!

$7K/month for the above infrastructure and a skilled team of 5.?

Cheap.

I look forward to seeing the proposal for the upgrades to services. I think given the cost relative to other services and if the projected run rate is reasonable (no more than double to $15K/month) go for funding for like 3-6 months (45-90K what 50-100HNY) and save some time for managing proposals and just manage the budget as Flora sees fit. Perhaps a reason to start building a reserve in xDAI is so that we don’t have to worry too much about HNY price fluctuating on budgets with longer time frames.

I will throw my HNY on this after we get pollen passed.

Thanks to everyone in Flora for their hard work and covering upfront costs for the community!

One minor suggestion. I know I helped pass a 1.3HNY proposal for @hernandoagf to fix the graph stuff. It would be nice to include in the Flora budget a bit for bounties to pay others. It is a bit easier in my mind to track outlays if they come through and are managed by Swarms, and bit easier on governance to have a better view of the value brought to the community by the appropriate swarms. Not a big deal just something to consider as 1Hive grows.

3 Likes

Gonna support it after gardens swarm proposal passed (still doesn’t have expected time)

Thank you @Eth_Man for your support and feedback! We will definitely take all of these into consideration for the next proposal. We decided to go with 2 months because of HNY fluctuation, but we thought about creating a proposal covering 3 or more months initially.

Also for the bounties, I completely agree. This proposal was a bit complex to get up due to all the factors and this swarm has been operating for longer than any other, so I kinda forgot about that possibility.

After the funds reach our Multisig, we are probably using Colony for easier payment management and paying potential bounties from there too. This is a 2-month proposal but we are not going to stop rewarding people that help because of this, if we ever use all of the funds before the estimated time we will create a new proposal with further details and probably with more accurate numbers.

1 Like

Sounds good. I am just thinking to save your group time and hassle on funding and just go for a longer time frame.

Thanks again @crisog for all the great work and thank you @Escanor for putting in the correct reference to @hernandoagf :smiley:

2 Likes