Madelyn Olson

The forkless approach at AWS does something slightly different but basically achieves the same effect. Instead of creating a new empty table, we maintain a a special type of iterator...

> I'm sad to hear that sometimes fork is still better. I would hope we could just replace fork with something strictly better. If it weren't for large collection data...

> I thought in that direction, but I'm not sure if there is an efficient way to process various edge commands that cover both keys located before the iterator and...

> Especially for scenarios such as Lua and modules. Lua scripts are supposed to pre-declare keys. I agree about modules though. We might have to fallback to CoW on the...

> Yeah, but in many cases they don't. It's a workaround that people use deliberately to get cross-slot transactions. Also things like what @artikell showed, but even just using ARGS...

> Do you just mean that the forked process would be multi-threaded to achieve higher throughput and reduce CoW simply by being faster? Yes. We can use multiple threads in...

@soloestoy @enjoy-binbin I would like to review the previous comment in the weekly meeting, so would appreciate if you could review it and post any comments here.

> https://www.vldb.org/pvldb/vol16/p1033-chen.pdf Are the improvements discussed in this paper in any Linux kernel?

@soloestoy I'm taking your comment to mean you would be "OK" with implementing this if we did, but you just don't think it's all that important. In the weekly meeting,...

I think it's a good quality of life improvement to Redis. @oranagra Pinging you since I would like your buy in before Jim does any more work. The current PR...