Managing database failover based on transaction request time

A method, comprising:

• copying a first dataset stored in a primary deployment to a secondary deployment to generate a second dataset;
• determining a first arrival time of a first request to perform a first transaction;
• determining a second arrival time of a second request to perform a second transaction
• executing the first transaction on the first dataset at the primary deployment in response to determining the first arrival time of the first request to perform the first transaction;
• and
• executing the second transaction on the second dataset at the secondary deployment in response to determining the second arrival time of the second request to perform the second transaction.

A non-transitory computer readable storage media comprising instructions that, when executed by a processor, cause the processor to:

• copy a first dataset stored in a primary deployment to a secondary deployment to generate a second dataset;
• determine a first arrival time of a first request to perform a first transaction;
• determine a second arrival time of a second request to perform a second transaction;
• execute the first transaction on the first dataset at the primary deployment in response to determining the first arrival time of the first request to perform the first transaction;
• and
• execute the second transaction on the second dataset at the secondary deployment in response to determining the second arrival time of the second request to perform the second transaction.