Device Utilization Threshold
Devices available for hire on io.net should remain idle and not perform any computationally intensive tasks. We are updating device requirements to ensure that CPU and GPU utilization rates do not exceed the thresholds specified in the tables below.
Starting November 4th, 2024 00:00:00 UTC, any devices that fail to adhere to the maximum utilization requirement will be considered non-compliant. Workers that fail to comply can be blocked and are subject to slashed Block Rewards.
GPUs
Supported Processor | Max CPU Utilization Threshold | Max GPU Utilization Threshold |
---|---|---|
A10 | 30% | 5% |
A100 80GB PCIe | 30% | 5% |
A100-PCIE-40GB | 30% | 5% |
A100-SXM4-80GB | 30% | 5% |
A10G | 30% | 5% |
A16 | 30% | 5% |
A40 | 30% | 5% |
GeForce GTX 1080 Ti | 30% | 20% |
GeForce RTX 2080 Ti | 30% | 20% |
GeForce RTX 3050 | 30% | 20% |
GeForce RTX 3060 | 30% | 20% |
GeForce RTX 3060 Ti | 30% | 20% |
GeForce RTX 3070 | 30% | 20% |
GeForce RTX 3070 Ti | 30% | 20% |
GeForce RTX 3080 | 30% | 20% |
GeForce RTX 3080 Ti | 30% | 20% |
GeForce RTX 3090 | 30% | 20% |
GeForce RTX 3090 Ti | 30% | 20% |
GeForce RTX 4060 | 30% | 20% |
GeForce RTX 4060 Ti | 30% | 20% |
GeForce RTX 4070 | 30% | 20% |
GeForce RTX 4070 Ti | 30% | 20% |
GeForce RTX 4080 | 30% | 20% |
GeForce RTX 4090 | 30% | 5% |
GeForce RTX 4090 D | 30% | 5% |
H100 80GB HBM3 | 30% | 5% |
H100 PCIe | 30% | 5% |
L4 | 30% | 5% |
L40 | 30% | 5% |
RTX 4000 | 30% | 20% |
RTX 4000 SFF Ada Generation | 30% | 20% |
RTX 5000 | 30% | 20% |
RTX 5000 Ada Generation | 30% | 5% |
RTX 6000 Ada Generation | 30% | 5% |
RTX A4000 | 30% | 20% |
RTX A5000 | 30% | 20% |
RTX A6000 | 30% | 5% |
Tesla T4 | 30% | 5% |
Tesla V100-PCIE-16GB | 30% | 5% |
Tesla V100-PCIE-32GB | 30% | 5% |
Tesla V100-SXM2-16GB | 30% | 5% |
Tesla V100-SXM2-32GB | 30% | 5% |
Tesla V100S-PCIE-32GB | 30% | 5% |
CPUs
Supported Processor | Max CPU Utilization Threshold |
---|---|
M2 Max | 30% |
M2 Pro | 30% |
M2 Ultra | 30% |
M3 | 30% |
M3 Max | 30% |
M3 Pro | 30% |
Updated 1 day ago