Speedy Transactions in Multicore In-Memory Databases

Re:Speedy Transactions in Multicore In-Memory Databases by WatchDog on Monday February 27, 2023 @07:55AM (#34955849)

So this design uses an epoch of 40 ms, does that mean any write operation will take on average 20ms?

Quite a lot of latency compared to regular databases.

[ Parent | on HN ]

Re:Speedy Transactions in Multicore In-Memory Databases by singron on Monday February 27, 2023 @10:54AM (#34957841)

I think so. Figure 7 is the only figure in the paper that shows latency, and it appears the epoch defines a latency floor. The system is aggressively optimized for throughput at the expense of commit latency, which is common for anything with group commit.

[ Parent | on HN ]