Filecoin Space Race: Miner Show & Tell
Below, we have condensed the presentation into an interview-style format for viewers to read and get a better understanding of who we are, and our plans for the future. Feel free to open up our powerpoint so you can reference the talking points along with the diagrams. We’ve also included the video of the actual event, released on Youtube for you to stream.
Let’s get started!
Q: Can we start off with a quick self-introduction?
A: Yes, of course! Hello everyone, my name is Neo Ge, I co-founded IPFSMain back in 2018, and we’re the service provider inside of the Filecoin ecosystem. I’m very happy to be here, also, thank you for having us. We’ve been committed to the Filecoin Project for almost 3 years now, and it’s been such an incredible journey for all of us in IPFSMain, and working with the Protocol Labs team has always been such a pleasure. It’s really exciting to present today. The presentation today is basically going to go through our Filecoin mining solution which includes Mining Architecture and Network Topology, then I’m gonna talk a little bit about our hardware information and our team as well.
Q: How has IPFSMain developed their filecoin mining architecture?
A: Over the past 2 years, we have had multiple architecture designs for large- scale Filecoin mining. After thousands of hours of testing, we’ve decided to use this Filecoin mining architecture. In this current mining solution, we are dividing major hardware into 3 different parts: the storage rigs, sealing rigs & SNARKs rigs. By separating the storage and computation apart, this Architecture gives us really good scalability and improves the hardware utilization efficiency on a tremendous scale. In this design, we want storage rigs to complete PoSTs (at least the first step of PoST – Generate Candidates) on their own, at the same time, we use Sealing Rig Cluster for seal_pre_commit_phase1 aka PC1, PC2 and C1, and SNARK Rig Cluster for C2 and window PoST. We modified and rewrote part of the code, and made some redundancy changes to improve the PoST success rate. By doing all that with this mining architecture, we’ve significantly improved the mining efficiency especially for P1 and C2. For example, we could manage to finish C2 under 10 mins, and there is still a lot of improvement we could do in the next few weeks.
Q: What does your network topology look like?
A: In the Network Topology, we are using two different ISPs for each WAN to increase the redundancy. The Communication core switch uses multiple switches with 10-Gigabit optical ports for stacking, which is mainly connected to lotus_node and lotus_miner. The Data core switch uses dual switches with 10-Gigabit optical ports for stacking, which is primarily connected to lotus_miner, storage, seal_worker & SNARK Cluster. And the Storage access switches have 4 10-Gigabit optical ports + 24 Gigabit electrical ports, duo-switch stacking of course. The 10-Gigabit optical ports are connected to Data aggregation switches, and Gigabit electrical ports are connected to storage. Each storage is connected with two network cables respectively to the two switches, and we did the dual network port binding to improve the storage network stability.
Q: What is your O&M platform?
A: Our O&M Platform, also known as FIL Manager, is part of the Filecoin mining solution. The FIL Manager includes CMDB, CI/CD, lots of monitoring systems and work order system. Please reference the powerpoint for a definitive list. The CMDB can help us automatically discover the host, get the host hardware information, and identify host failures. The CI/CD module can help us perform fast and accurate deployment whenever there is an update.The business monitoring system provides comprehensive monitoring of the health status of the business from different perspectives of lotus, miner, seal,storage, snark, window PoST, and so on.
Q: Can you run us through the various specific monitors in your business monitoring system?
A: Yes, so we have a total of 7 different specialized monitors: the Lotus monitor, Miner monitor, Seal monitor, Storage monitor, SNARK monitor, Window PoST monitor, and Host monitor. The Lotus monitor helps us on node synchronization and the Miner monitor helps us on order status, miner status, and sector life cycle tracking.The seal monitor focuses on the working status of seal workers, like what is the sector ID and what stage that sector is currently in. It is able to identify various errors that occur during the sealing process and deal with faults automatically. It will stop receiving new tasks until the fault is repaired. The storage monitor helps us on the sector storage status, it will find storage failures actively, perform data migration in real-time, and make the migration history traceable. Snark monitor focuses on the working status of snark and the resource demand of it. Window PoST monitor is responsible for window post working status and history of all miners. And lastly the Host Monitoring Interface shows all the monitoring systems so you work together to identify and classify faults in hardware, services, and networks, and notify the corresponding staff to deal with them in time.
Q: How else does IPFSMain check and manage the workflow and ensure a constant streamlined operation?
A: Great question! So our work order system manages the workflow and can track various operations and maintenance events. All faults identified through the monitoring system can automatically generate work orders and dispatch them to the corresponding person in charge. We also utilize an exception handling system that catches all exceptions through logs and notifies O&M personnel to deal with it. If there are any advanced exceptions like synchronization exception, windowPoST exception, the system will call the corresponding O&M personnel through the phone interface and ask them nicely to deal with the exception. Pretty sweet, huh?
Q: Run us through a basic explanation of your hardware, and what a good mining solution means to you.
A: I believe designing a good mining solution is like building a wooden bucket. No matter how high a bucket is, the capacity of the bucket depends on the shortest plank. Filecoin mining is not like any other traditional computation cryptominings like Bitcoin or Ethereum. Filecoin mining is complex, especially when you are mining at a meaningful scale. If you are sealing too fast or too slow so that other processes have to wait and stay idle, then you’re probably wasting the hardware performance. So the key is to find a balance with cost-effective hardware. Apparently, cost-effectiveness or cost-efficiency is another metric of Filecoin mining solution, where you could get the most beefy machines regardless of cost, or find the right hardware for best cost-efficiency which also leads to a best Return on Investment ratio. It’s like running an Uber using Lamborghini versus say Honda Civic. Another thing I’d like to highlight is that we are ready for algorithm upgrades from SDR to NSE. That's actually one of the reasons why we chose current architecture and hardware designs, they are built for algorithm upgrades. I believe in order to make Filecoin a successful project is to invite everyone in the world to do the Filecoin mining, and to make it happen, we need to lower the hardware requirements, so that everyone could mine at any scale.
Q: Are you also actively participating in the Filecoin Space Race right now?
A: Yes! We are 4 days into the Space Race, and even though we are getting attacked from time to time, our team has been taking appropriate actions and conducting different types of defenses. I think one of our nodes – Miner ID is t01235, is ranking number 5 on the board, and we also have multiple nodes running for the Space Race, and with all these miners combined it's safe to say that we are ranking top of the board not only in the Power category, but also in the Mining Efficiency. Now we are having over 80 people hard-working every day not just for this Space Race, but also for the coming mainnet launch.