vkazanov

45 ms·

631.
by vkazanov 7y ago
...ok, then what are the alternatives? And why is it so bad? Something that can handle hundreds of terabytes on hundreds of machines and provides useful tools on top of the whole thing (Spark, Hive, etc)?
632.
by vkazanov 7y ago
I am not a certified specialist in Hadoop-related things but there's definitely an industry of people working with Hadoop/HDFS/Spark/etc. It's almost a standard in Business Intelligence where I work these days. So t
633.
by vkazanov 7y ago
I'll correct you: Hadoop solves a hard problem that lots of big enterprises have. A distributed file system is a big deal, and everything that on top also is. Now making (big)money with its ecosystem is another question.
634.
by vkazanov 7y ago
CPython is a relatively simple program calling for a (relatively) simple language. It's meant to be understood by lots of people, used within many different contexts and C is in much better position here. At its heart it is just a byte
635.
by vkazanov 7y ago
Yes, that. I find it really... Wonderful how types and declarations are inferred from actual code using 'em. OTOH, this is something I see a lot in Hindley-Milner typed languages, although they don't usually go too far here, i.e.
636.
by vkazanov 7y ago
I tend to just use the Dell XPS with Ubuntu preinstalled: hardware-related things really do just work. For everything else I have Emacs :-)
637.
by vkazanov 7y ago
I wouldn't be so sure. This thing given a syntactically valid piece of code infers all declarations required by it. They implemented a working demo, along with some helper code so they are trying to push it. How about IDEs?
638.
Psyche-c – Automatic Compilation of Partially-Available C Sources(cuda.dcc.ufmg.br)
37 points by vkazanov 7y ago|6 comments
639.
Regex bytecode interpreter: looking for needles in session haystacks(badootech.badoo.com)
1 points by vkazanov 8y ago|0 comments
640.
Regex bytecode interpreter: looking for needles in session haystacks(badootech.badoo.com)
1 points by vkazanov 8y ago|0 comments
641.
by vkazanov 8y ago
Well, NOW I am curious. And I do a lot data engineering these days ;-) btw, do you implement any edgeql-specific optimizations? I know that the db shares most of the code with psql, and its optimizer is okay. But being closer to the math fo
642.
by vkazanov 8y ago
So, EdgeQL is closer (than SQL) to relational algebra as it was originally envisioned? Did Tutorial D have any influence on the language and the system?
643.
by vkazanov 8y ago
My dad's in his 60s. Runs marathons once a year. Works on backend stuff in a ML startup. I am not sure I can make 10 km these days...
644.
by vkazanov 8y ago
> I still don't understand why they felt like they had to add half-baked classes / OOP to it. That one I'll never forgive :-)
645.
by vkazanov 8y ago
I always strongly disliked the OOP school of program design, even though I learned to parrot everything required by one's typical interview process. Keeping data and functions together can be nice. In some rare cases inheritance is a
646.
by vkazanov 8y ago
I don't really know if 33 is young or old but I really cannot agree more about the mantra thing. And it's not only the OOP craze that completely ate the industry at some point..! NoSQL-related discussions, all UIs and CLIs, the bi
647.
by vkazanov 8y ago
When you say "genius" you mean his technical abilities, by "unprofessional" you probably mean those moments when he pointed out other people's technical mistakes in a... straightforward manner. Some people are offen
648.
by vkazanov 8y ago
Sounds like a very interesting project! Could you share some details? What techniques did u use? I collect everything interpreter-related. But that's easy to see from the article :-)
649.
by vkazanov 8y ago
Yeah, it's a very good collection you got there! some of the papers I read, others are in the queue. I was also surprised to see the switch-based solution winning here. But I was even more surprised to see how a simple change (a prim
650.
by vkazanov 8y ago
What's so surprising? Jit-like techniques are older than I am, and I am not that young anymore :-)
651.
by vkazanov 8y ago
Well, this is the perfect case for jits: basic number crunching. Bytecode interpreters are pure overhead in this case. Problems usually come from jumping between jitted/non-jitted code or, say, pointer chasing in complex objects. Did y
652.
by vkazanov 8y ago
Interestingly trace caches are now gone in both hardware CPUs and major jit compilers :-) Correct me if I am wrong..?
653.
by vkazanov 8y ago
So you did a wrapper around libjit and generated functions on-the-fly? I am not sure it counts as a full bytecode jit. :-) I did a very stupid jit for the PigletVM thing while working on the article and the difference was more like 1.2-1.5
654.
by vkazanov 8y ago
That's very true. In fact, I took one of the techniques (traces) directly from a paper describing how the Bochs x86 virtual machine works :-) But it goes deeper than that. I believe the whole "trace" thing in both jits and vm
655.
by vkazanov 8y ago
I am the author of the article :-) This sentence is more like a description of the current state of affairs: people go for very complicated jits instead of trying to push the basic interpreter to its limits. And jits are complicated , and
656.
by vkazanov 8y ago
Source, pls. Where and when did he say that?
657.
by vkazanov 8y ago
I think I only saw DP-based algorithms 3 times in 12 years of working and reading open source code. Two of them were boring textbook-based implementations. One algorithm was original, and came from 80s - it's from Emacs source code, wi
658.
by vkazanov 8y ago
Having no threads is exactly what makes Redis so convenient in deploying it to clusters: you have a node, and the node is a single process. It's like a building block. Also, Redis source code is so much cleaner than memcached!
659.
When Pigs Fly: Optimising Bytecode Interpreters(badootech.badoo.com)
2 points by vkazanov 8y ago|0 comments
660.
Home-grown bytecode interpreters(badootech.badoo.com)
1 points by vkazanov 8y ago|0 comments

More ›