Switching of workloads according to switchover cost
A method for automated switching of workloads on computing devices to increase returns such as rewards or transaction fees in a cryptocurrency blockchain network is disclosed. A plurality of signals impacting the profitability of mining for a plurality of different cryptocurrencies and plurality of different mining pools are monitored. In response to the plurality of signals indicating a different cryptocurrency and mining pool combination is more profitable, the computing device workload is automatically switched. Switching cost may be calculated and used to prevent unprofitable switches. The signals may be used to train a machine learning model that may be used to predict future profitability for automatic switching.
1. A method for optimizing returns generated by a computing device, the method comprising:
instructing the computing device to work on a first workload, wherein the first workload is mining a first cryptocurrency;
monitoring a plurality of signals impacting a profitability of mining for a plurality of different cryptocurrencies; and
in response to the plurality of signals indicating a second cryptocurrency is the most profitable of the plurality of different cryptocurrencies, calculating a difference in profitability from the first workload being worked on by the computing device and a second workload being worked on by the computing device where the second workload is mining the second cryptocurrency; calculating a switchover cost representative of a time period during which no work is performed by the computing device when changing from the first workload to the second workload; and automatically switching the computing device to mine the second cryptocurrency if the difference in profitability is greater than the switchover cost, and mining the second cryptocurrency.
2. The method of claim 1 , wherein the plurality of signals comprises mining difficulty, exchange rates, and pool fees.
3. The method of claim 2 , wherein the plurality of signals further comprises difficulties, hashrates, and power costs.
4. The method of claim 3 , wherein the monitoring is performed in real-time.
5. The method of claim 1 , further comprising: measuring a switching time for the computing device; and using the measured switching time to calculate the switchover cost.
6. A method for optimizing returns generated by a computing device, the method comprising:
monitoring a plurality of signals impacting mining profitability for a plurality of combinations of different cryptocurrencies and different mining pools;
calculating a most profitable cryptocurrency and mining pool combination from the plurality of combinations of different cryptocurrencies and different mining pools;
instructing the computing device to switch workloads to the most profitable cryptocurrency and mining pool combination, and mining the most profitable cryptocurrency;
calculating a difference in profitability from a current workload being worked on by the computing device and the most profitable cryptocurrency and mining pool combination; and
calculating a switchover cost representative of a time period during which no work is performed by the computing device when changing workloads, wherein the instructing is only performed if the difference in profitability is greater than the switchover cost.
7. The method of claim 6 , wherein the plurality of signals further comprises difficulties, hashrates, and power costs.
8. The method of claim 6 , wherein the monitoring and the calculating are performed on a management computer that communicates with the computing device via a network.
9. The method of claim 6 , wherein the monitoring and the calculating are performed in real time.
10. The method of claim 6 , wherein the plurality of signals are selected from a group comprising: expected reward, current difficulty, current exchange rate relative to a fiat currency, the computing device's hash rate, the computing device's power consumption, and the computing device's predicted failure rate.
11. The method of claim 6 , wherein the plurality of signals comprises mining difficulty, exchange rates, and pool fees.
12. The method of claim 6 , further comprising: training a machine learning model using historical data of the plurality of signals to identify correlations between the plurality of signals and the profitability of each of the plurality of cryptocurrency and mining pool combinations; and using the trained machine learning model to perform the calculating of the most profitable cryptocurrency and mining pool combination.
13. The method of claim 6 , further comprising: training a machine learning model based on historical data to identify correlations between the plurality of signals and the profitability of each of the plurality of cryptocurrency and mining pool combinations; and using the trained machine learning model to determine when to perform the instructing and a power level setting at which to operate the computing device.
14. A non-transitory, computer-readable storage medium storing instructions executable by a processor of a computing device, which when executed cause the computing device to:
monitor a plurality of signals impacting a profitability of mining for a plurality of different cryptocurrencies in a plurality of different mining pools;
calculate a most profitable cryptocurrency and mining pool combination based on the plurality of signals;
calculate a switching cost;
predict a benefit to switching to the most profitable cryptocurrency and mining pool combination; and
switch workloads to the most profitable cryptocurrency and mining pool combination if the switching cost is less than the predicted benefit, and mining the most profitable cryptocurrency.
15. The non-transitory, computer-readable storage medium of claim 14 , wherein the plurality of signals further comprises difficulties, hashrates, and power costs.
16. The non-transitory, computer-readable storage medium of claim 14 , wherein the instructions are firmware executable by the computing device, wherein the computing device executes the workloads.
17. The non-transitory, computer-readable storage medium of claim 14 , wherein the computing device is a management computer, wherein the instructions are an application executable by the management computer, wherein the management computer manages another computing device that executes the workloads.
18. The non-transitory, computer-readable storage medium of claim 14 , wherein the computing device is a management computer, wherein the instructions are an application executable by the management computer, wherein the management computer manages a plurality of computing devices that execute the workloads.
19. The non-transitory, computer-readable storage medium of claim 14 , wherein the instructions are firmware executable to optimize a return on investment for the computing device.
20. The non-transitory, computer-readable storage medium of claim 14 , wherein the plurality of signals comprise network hash rate, the computing device's power consumption, the computing device's hash rate, the power cost for the computing device, the computing device's estimated lifespan, the computing device's estimated failure rate, and the computing device's estimated switchover cost.