I wonder when this "databases are nearly always bottlenecked on IO" perception finally is going to die. I think it's been false for > 50% of instances for at least 15 years. And it's just plainly wrong when we can have small-ish servers with >16 internal NVMe drives.
Even if you don't, you have access to high bandwidth, very parallel, but higher latency storage.
-
-
Or not high bandwidth either. Sure, you can have high bw, but not everybody pays for that. The norm I'd say is not super fast disks, hence I/O still being the bottleneck.
-
Sure, those exist. I didn't say anything to the contrary? Even if you have slow-ish disks, most of the time you still have enough memory to keep most of the hot data set in memory.
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.