vemv

14 ms·

151.
by vemv 5y ago
As long as there's more than 1 thread processing the queue, the queue is being concurrently processed. Particularly, the queue items will be picked up in a nondeterministic order, driven by CPU utilization. > Is the idea here that y
152.
by vemv 5y ago
A large monolithic app with all sorts of workloads going on in it seems hard to reason about, measure, GC-tune, HA, etc in the first place. So a modern choice would be either to have specialized background job processes, or microservices. E
153.
by vemv 5y ago
An informed thread-based approach is quite more careful than just spinning "lots of threads" and can certainly achieve a concurrency of 50, or 10000. Let's say we want a concurrency of 1000 and have 8 CPUs available. I'd
154.
by vemv 5y ago
IO doesn't cause the whole CPU to be idle. It simply causes a context switch, which will allow another thread to do work.
155.
by vemv 5y ago
This seems the sort of problem that is well-solved by a pipeline pattern. Each member of the pipeline is a thread pool that does exactly one type of job. One can allocate thread pools sensibly relative to CPU count and achieve ideal task&lt
156.
by vemv 5y ago
> For example, the simplest way to write a network server that can handle more than one client at a time is to fork a new thread for every client. This seems much debatable. For me the simplest way is creating a thread-per-core pool, and
157.
by vemv 5y ago
> On the JVM, native threads are really expensive, especially in terms of memory. I don't think this is true if you create have approximately 1 thread for each CPU core (which is the ideal scenario in a non-async app - why would you
158.
by vemv 5y ago
Genuine question - is there a concrete and practical problem that can't be solved without Loom green threads on the JVM? Generally I think that programs can be structured nicely and have them perform optimally by using traditional thre
159.
by vemv 5y ago
I'd suggest to change the editorialized submission title, as it predisposes people to share opinions that can easily be unrelated to the article's contents in question.
160.
by vemv 5y ago
> Test generated by Diffblue Deeptest. I think this is the reason for that duplication and numbering
161.
by vemv 5y ago
I've observed from my personal job experiences that Eastern European developers tend to pay more attention to detail. It's a double-edged sword as it can yield more correct code but also spend time in things that a formatter/
162.
by vemv 5y ago
Focus on what you have to offer to companies. Everything else is secondary. - What's your role? (dev, QA) - What's your speciality? (python, mobile, etc) - Are you good at your role and speciality? How can you show so? (I complete
163.
by vemv 5y ago
VAR is just people, it's not like an algorithm is dictating the results.
164.
by vemv 5y ago
Lately I was thinking that probably something like 80% of what developers really seek in a solution in this space, whether it's based on k8s, Heroku, serverless, VPS, Chef, Terraform, whatever, is the deployment tool/experience.
165.
by vemv 6y ago
I cannot make an expert assessment myself but for one thing mysql is occasionally picked precisely because of its HA story. It was the case in a couple past jobs of mine, back then I didn't particularly appreciate it but now I might se
166.
by vemv 6y ago
Like many developers, I've used postgresql unquestioningly for many years, say 10. But if you ask me, I've rarely have had to face stringent scaling or availability requirements. Many deployments were a variation of AWS RDS or a s
167.
by vemv 6y ago
I'm shamelessly passionate about software development, but I don't project it being a big part of my life forever and ever. For example with covid, I have fewer other possible hobbies so doubling down in this specific interest is
168.
by vemv 6y ago
This is a good example of why anything related with threading makes me nervous when using Ruby. When a language is mutable and thread-unsafe by default, keeping the whole stack sane at all times seems prohibitively arduous. One cannot code-
169.
by vemv 6y ago
It would not trigger the same bug. But since it's a multi-threaded server, it will exercise arbitrary code (Rails core, your app, depended-on gems) in a concurrent manner. This can unmask race conditions that wouldn't manifest the
170.
by vemv 6y ago
Clojure is a good option as well - it can be seen as FP without the type theory - i.e. a more distilled version. It also sports a pragmatic community - web, CLI, UI etc utilities abound.
171.
by vemv 6y ago
That might be also considered a multirepo in disguise :)
172.
by vemv 6y ago
How about: business logic is and remains private, but support libraries, boilerplate, etc is offered as discrete OSS repositories. This isn't too fancy - quite commonly corps and startups alike offer some generic libraries. One can tak
173.
by vemv 6y ago
I also have a substantial piece of Rubocop config that I use for my own projects. Why is anyone else's better than mine, or vice versa? Why do have people have to name their preferences 'standard' and promote them around? Suc
174.
by vemv 6y ago
I wonder if the quite extremely early/intense competition that Roam faced as soon as it was shared publicly will result in none of the competitors succeeding (at the very least as a viable for-profit company). It's possible that
175.
by vemv 6y ago
Related, I really wish threads had a tooltip with a tldr of its contents. Sometimes it's impossible to tell what the article talks about without clicking on it. Which can be subtly exploited by using clickbaity titles, or 'fixed&#
176.
by vemv 6y ago
Modular First. You start out as technically a monolith, but that is prepared at all times to be decomposed into services, if and when the need arises. It's nothing too fancy - can be simply another name for Hexagonal Architecture, func
177.
by vemv 6y ago
Personally I will always applaud little experiments that question the status quo and that may be fun for those involved. What's the worst thing that could happen?
178.
by vemv 6y ago
Disclosing the conditions doesn't seem fortunate on their side.
179.
by vemv 6y ago
> The wrists themselves should not be under strain or pressure, including wrist-rests > The hands should move freely over the keyboard > The idea of "less movement" is incorrect, and instead I should aim for "less st
180.
by vemv 6y ago
I also count myself between those who implemented such a script when reading about github1s on HN :) My latest iteration involves not using the terminal at all - I open the project from emacs instead. So I `M-x peek-repo`, paste the url, an

More ›