I came up with a clever one recently. So, I had a stream of records of fixed size where I will receive exactly one per second. I needed to keep approximately tw

I came up with a clever one recently. So, I had a stream of records of fixed size where I will receive exactly one per second. I needed to keep approximately two weeks of this data. I would also need to be able to query the records by time. Anyway, the idea is: use modulus on the file creation time to calculate a write position. Use the location of the current write position to locate query starts and stops. The file then has zero space overhead. Is this something known or did I actually invent something here? If this is my invention I'm open to hearing suggestions for a cool name!