Thanks for the comprehensive report and all the hard work that went into it! I’ve got a few questions to clarify some technical details:
Hive - Node Questions:
- Could you please provide more details on how you're running the Hive nodes? Specifically, what Docker image are you using? Are these full nodes? If possible, could you share the docker-compose file or some documentation to help us better understand the setup?
- Are there any special configurations needed to run the validator network on Hive nodes, or can we use any public Hive node for this purpose?
Proxy & Caching:
From what I understand, SPL runs its own Hive nodes that are cached through Cloudflare, which is then used by the game. Will this setup continue to function the same way when the validator network goes live? Additionally, will the costs associated with caching and implementing Cloudflare and R2 Storage continue to be covered by SPL? If so, is this a significant cost factor we should consider?
sl-validator Network:
Could you also clarify the purpose of the AWS account? Will the SPS-chain be able to run independently from AWS and Terraform, or are these dependencies essential?
Looking forward to your insights!
We can share the Dockerfile with you later this week, but to give you a quick summary: our setup is designed to simplify the management of Hive nodes, making it easier to run, reset, delete, or archive nodes. The downside is that it requires a significant amount of storage—about 4TB of NVMe with ZFS.
No special configurations are required. The validator network uses standard Hive API calls, so any public Hive node will work just fine.
Yes, the setup will continue to function the same way when the validator network goes live. As for costs, they are currently covered by SPL, with expenses around $3,000 per month due to the contract. However, the actual runtime costs are lower and probably closer to $1,000 per month. Splinterlands will try to keep it running as a service to the community and to our organization.
The AWS account is currently being used to verify that the validator integrates seamlessly with the existing SteemMonsters API. At this stage, it's primarily for testing. Eventually, the validator will transition to the production account, and the current setup will be phased out. The SPS chain is designed to operate independently of any specific instance of the validator network, including Splinterlands, which will function within a decentralized environment. However, we plan to implement a fallback system where transactions are stored independently in a centralized database for tracking, roll-ups, and our own analytics.
Thanks, PJ, for your response. It really helped to clarify some of my questions.
Yes, that would be incredibly helpful. It would allow me to better understand how everything works and potentially set up a node myself that aligns with the Splinterlands infrastructure.
I’m looking forward to discussing more details like this. I believe it’s crucial to stay engaged with the development within the community so we’re fully prepared when it goes live. I'm still a bit unclear on how or which API calls will be handled by the validator software, but I’m confident we’ll figure it out as we go.