Uptime
Uptime is the share of time an asic ran and sent solutions to the pool. It is counted as a percentage of a period: over a day, a week, a month.
What uptime is
An ASIC is either hashing or it is not, and there is little in between. Uptime shows how much of the time it was working. Ninety nine percent over a month sounds pretty, and in practice that is seven hours of standing still, which is roughly a third of one day's revenue.
Judging uptime by the lights on the case makes no sense, because an asic can roar and heat the room while its link to the pool is dead. Judge it by the pool: it sees the shares that arrive and it knows when the worker went quiet. That is why pool stats carry an activity chart, and that chart is more honest than anything the miner reports about itself.
Typical causes of downtime come in a clear order. First place goes to the network: the router, the cable, a switch port. Second place is heat and thermal protection. Third is a dead power supply, and fourth is a hashboard, after which the miner keeps running with a third of its hashrate gone.
Quick reference table
| What it shows | the share of time in operation, as a percentage |
| Where to look | the worker activity chart in the pool |
| 99 percent over a month | about 7 hours of downtime |
| 95 percent over a month | about 36 hours of downtime |
| Common cause | network first, then heat, then the power supply |
How to raise uptime without spending money
Turn on pool alerts for a quiet worker with a threshold of about fifteen minutes. It costs nothing and it turns a day of downtime into twenty minutes. We see the difference across clients: those who set up alerts measure downtime in hours, while everyone else measures it in days.
Put three pool addresses into the miner, a primary one and two backups. Half of all so called freezes are really one pool being unreachable, while an asic with a single address just sits and waits.
And keep a spare power supply on the shelf. In our shop it is the part that fails most often, and downtime spent waiting for a delivery eats more than the spare costs.
An example in money
An Antminer S21 eats 84 kWh a day, and on the Spanish rate that is about 11 euros of expense per day. Every day of downtime means one day of revenue lost, and the electricity you saved that day does not cover it. Hence a simple rule: chase the first percent of uptime rather than the last percent of hashrate.
Related terms
Where to go next on the site
Hardware
- Antminer S21 tool
- ASIC Miner knowledge base
Numbers
Questions and answers
What uptime counts as normal
On a home site 98 or 99 percent is reachable without spending anything, and past that you are fighting for single hours. Industrial sites with redundancy post higher figures, and they pay for them in infrastructure.
Why does the pool show downtime when the asic was running
The pool counts the shares that arrive. If the link dropped, the miner spins its fans and heats the room for nothing, and the work goes nowhere. For your income that is the same as downtime.
Does one hour of downtime really hurt income
An hour is about four percent of a day. Over a month a couple of those go unnoticed, while a single night nobody spotted is already a third of a day's revenue.
How the terms connect
Every link in the chain is clickable. Orange marks where you are now.
Looking for an ASIC miner
The catalog holds 212 models. You can compare them by hashrate and by joules per terahash, then plug your own rate into the calculator and see what stays in your pocket.
Page written and checked by Denys Klimchuk. Updated .