tkhattra

one thing to be aware of is that duckdb doesn't currently allow concurrent readers and writers (unlike sqlite). depending on your use case, that might be a showstopper.

[1] is a particular one regarding specification languages and types. its background description about a "rabidly pro-types" referee is quite interesting...

[1] http://lamport.azurewebsites.net/pubs/pubs.html#lamport-type...

the paper mentions that CLP comes with a default set of schemas, but you can also provide your own rules for better compression and faster search

sorry, i'm not familiar with Sol. might be worth inquiring on the lua mailing list

i don't think that's true - lua stores sparse arrays efficiently. the table above should only have 2 slots allocated. see [1] or [4] for details on how lua implements sparse arrays.

[1] https://cacm.acm.org/magazines/2018/11/232214-a-look-at-the-... [2] http://www.lua.org/doc/jucs05.pdf

i think Peter Weinberger is still writing code at google...

In the Erlang VM, each Erlang "process" has its own garbage collected heap [1]. There's also an ETS module for more efficiently storing and sharing large amounts of data, but data stored in ETS is not automatically garbage collected [2].

[1] https://www.erlang.org/doc/apps/erts/garbagecollection [2] https://www.erlang.org/doc/man/ets.html

see this 1995 Wired interview with ken thompson - https://www.wired.com/1995/08/thompson-4/.

pcstat and vmtouch (page cache statistics and control)

https://github.com/tobert/pcstat https://hoytech.com/vmtouch/

Re: "proving it works", in a previous interview Lamport said that some folks measure simplicity by showing something to people and they say whether or not they understand it, while he measures simplicity by showing something to people and they can prove it's correct.

c2c stands for cache-2-cache.

perf-c2c allows you to measure cacheline contention - see https://man7.org/linux/man-pages/man1/perf-c2c.1.html and https://joemario.github.io/blog/2016/09/01/c2c-blog/.

TLI/XTI are from AT&T SVR3/SVR4, not Research Unix (V8, V9, V10). dmr's V8 streams were the basis for AT&T SVR3/SVR4 STREAMS (all caps!), but quite a bit simpler: https://en.wikipedia.org/wiki/STREAMS.

just make sure it's not "overly optimized" - https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95189 (in this case it was gcc's builtin memcmp that was broken, not glibc's) :)

Hardware and Software Support for Virtualization Synthesis Lectures on Computer Architecture (2017)

https://www.morganclaypool.com/doi/abs/10.2200/S00754ED1V01Y...

Bringing Virtualization to the x86 Architecture with the Original VMware Workstation (2012)

https://dl.acm.org/doi/abs/10.1145/2382553.2382554

yes, set also. go's map implementation uses this technique - allocate a new array of buckets and gradually migrate from old to new buckets (i think it does the migration during set and delete only, not get).

I think it was Mike Burrows, but I can't remember where I heard or read that.

The Elements of Networking Style, by Michael Padlipsky - an (extremely!) opinionated critique of the OSI model vs the Arpanet (TCP/IP) model.