Validator Assignment Strategy Guide
Overview​
This guide explains the recommended strategy for assigning your GPU executors to validators on the Lium Compute Subnet. The strategy is designed to maximize your mining rewards based on how validator stake influences reward distribution in the Bittensor network.
Understanding the Fundamentals​
How Rewards Work​
In Bittensor subnets:
- Validators score miners - Each validator evaluates miner performance independently
- Stake = Influence - A validator's influence on your rewards is proportional to their stake
- Weighted consensus - Your final rewards are determined by the weighted average of all validator scores
If a validator holds 40% of the network stake, their evaluation of your work accounts for approximately 40% of your final reward calculation.
Why Assignment Strategy Matters​
Consider this scenario:
- You have 10 executors
- Validator A has 40% stake but you only assign 1 executor to them
- Validator D has 5% stake but you assign 3 executors to them
Result: You're over-serving a low-influence validator and under-serving a high-influence validator, reducing your potential rewards.
The Stake-Weighted Assignment Strategy​
Step 1: Analyze Validator Stakes​
Visit TaoMarketCap Subnet 51 Validators to see current stake distribution.
Look for:
- Stake percentage - Each validator's proportion of total network stake
- Active status - Ensure validators are currently operational
- Hotkey addresses - The SS58 addresses for assignment
Step 2: Ensure Coverage of Major Validators​
The original strategy suggests covering validators that control "at least 50% of the total network stake." This recommendation comes from the idea that:
- You want to serve validators who collectively have enough influence to significantly impact your rewards
- Focusing only on tiny validators (e.g., those with less than 1% stake each) means missing the validators who have the most say in your earnings
The "50% threshold" is not a hard rule from the protocol, but rather a practical guideline. The actual optimal coverage depends on:
- How many executors you have
- The current stake distribution
- Validator activity patterns
Step 3: Apply Stake-Weighted Distribution​
Once you've identified which validators to serve, distribute your executors proportionally to their stake.
Example with 100 Executors​
Given this stake distribution:
| Validator | Stake % | Description |
|---|---|---|
| Validator 1 | 50% | Dominant validator |
| Validator 2 | 25% | Major validator |
| Validator 3 | 15% | Significant validator |
| Validator 4 | 5% | Minor validator |
| Validator 5 | 5% | Minor validator |
Recommended distribution:
- Validator 1: 50 executors (50% of your capacity)
- Validator 2: 25 executors (25% of your capacity)
- Validator 3: 15 executors (15% of your capacity)
- Validator 4: 5 executors (5% of your capacity)
- Validator 5: 5 executors (5% of your capacity)
Example with Limited Executors (10 GPUs)​
With fewer executors, you need to be more selective:
| Validator | Stake % | Executors | Reasoning |
|---|---|---|---|
| Validator 1 | 50% | 5 | Half your capacity to the most influential |
| Validator 2 | 25% | 3 | Roughly proportional to stake |
| Validator 3 | 15% | 2 | Cover third-largest validator |
| Validator 4 | 5% | 0 | Skip smaller validators with limited resources |
| Validator 5 | 5% | 0 | Skip smaller validators with limited resources |
Step 4: Implementation​
For each validator you want to serve:
# Add executor to validator
docker exec <miner_container> python /root/app/src/cli.py add-executor \
--address <executor_ip> \
--port 8001 \
--validator <validator_hotkey>
Step 5: Monitor and Adjust​
Validator stakes change over time due to:
- Delegators moving their stake
- New validators joining
- Existing validators leaving
Recommended monitoring schedule:
- Weekly: Check for major stake shifts (greater than 10% change)
- Monthly: Comprehensive review and rebalancing
- Immediately: If you notice significant reward changes
Practical Considerations​
When to Deviate from Pure Stake-Weighting​
Consider adjusting the strategy if:
- Validator specialization: Some validators might prefer specific GPU types
- Geographic distribution: Network latency might favor certain validator-executor pairs
- Request patterns: Some validators might be more active than others
- Minimum service requirements: Some validators might require a minimum number of executors
Real-World Constraints​
The stake-weighted strategy assumes ideal conditions. In practice:
- Validators may have downtime - Monitor which validators are actively sending requests
- Network conditions vary - High latency to certain validators might reduce effective performance
- Not all work is equal - Some validators might send more computationally intensive tasks
Common Scenarios​
Scenario 1: New Miner with 1-3 GPUs​
Approach: Focus on the top 1-2 validators
- Assign all executors to the highest-stake validators
- Monitor performance before expanding
Scenario 2: Growing Operation (10-20 GPUs)​
Approach: Implement proportional distribution
- Cover top validators representing 70-80% of stake
- Use stake-weighted distribution
- Keep 1-2 executors flexible for testing
Scenario 3: Large Operation (50+ GPUs)​
Approach: Full strategic distribution
- Cover all significant validators (above 1% stake)
- Maintain strict proportional distribution
- Reserve 5-10% capacity for emerging validators
- Implement automated rebalancing
Troubleshooting​
Issue: No rewards despite following strategy​
Possible causes:
- Validators not actively requesting work
- Network connectivity issues
- Executor performance problems
- Recent stake distribution changes
Solutions:
- Verify validators are active using network explorers
- Check firewall and network configurations
- Monitor executor logs for errors
- Review recent stake changes and adjust
Issue: Lower rewards than expected​
Check:
- Are your executors actually receiving and completing requests?
- Has the stake distribution changed significantly?
- Are other miners offering better performance?
Summary​
The stake-weighted assignment strategy is based on the principle that your rewards are influenced by validators in proportion to their stake. While the "50% coverage" and proportional distribution guidelines are not hard protocol rules, they represent a logical approach to maximizing rewards given how the Bittensor consensus mechanism works.
Remember:
- This strategy is a starting point, not a guarantee
- Monitor actual performance and adjust based on real results
- Stay engaged with the community to learn from collective experience
- Be prepared to adapt as the network evolves
Additional Resources​
- Live Validator Data: TaoMarketCap Subnet 51
- Community Support: Discord Channel
- Technical Issues: GitHub Repository
Note: This strategy is based on community observations and logical analysis of the Bittensor reward mechanism. Actual results may vary based on network conditions and implementation details.