Robert Muir
Robert Muir
I'm pretty sure it boils down to a compile-time macro. For arch linux on mike's machine with many cores, easiest solution is to switch to `linux-lts` kernel "with configuration options...
note the guava version shenanigans... i'd have to test, but i don't think any of the new 3 checks apply to us... two are for frameworks not in use. The...
profilers often lie about this. I think we should keep the synchronization simple. the IDs are important to not duplicate.
this ID is generated for safety, and we only calculate this **once per segment** when initially flushing it. Doesn't even happen at merge. Really, it should be a few insns...
This results in a lot more code complexity, which makes maintenance difficult. Maybe the version of java you are testing with has a bug in its register allocator or something?...
I don't know the impact it would take on benchmark times, but I know this is already substantial, defer to Mike on that. Obviously a (potentially huge?) part of the...
maybe at least we could run just one or two queries with small top-N to make sure that WAND-type optos still work? not sure how hard it would be, but...
IMO: just use a filesystem with this feature such as zfs.
The advantage of letting a filesystem such as zfs (which was designed to do exactly this), is that it is integrated in the correct place and operating system caches work...
@easyice Something like DELTA+FOR shouldn't require any cache, right? To me that is a different problem with other challenges: index would need to be e.g. sorted on timestamp field for...