21 ms·

1,000,000 requests in 52 minutes is just 320 req/sec. Am I missing something? What's so amazing about this? I just deployed some production feed that serves a

1,000,000 requests in 52 minutes is just 320 req/sec. Am I missing something? What's so amazing about this? I just deployed some production feed that serves at 1955 requests/second on a cheap VPS in freaking PHP, one of the slowest languages out there.

aaossa 11y ago
Why you say is not amazing? Honestly curious here :)
- imaginenore 11y ago
  Because it's trivial. I would be interested in anything doing 10,000+ req/sec on a cheap VPS. 320 is nothing. People achieve 2 million requests/second with C++ on EC2: https://medium.com/swlh/starting-a-tech-startup-with-c-6b5d5856e6de#.38xs3etwg https://medium.com/swlh/starting-a-tech-startup-with-c-6b5d5...
  - aaossa 11y ago
    Oh I see now... This speaks by it self: C++/Proxygen =1,990,130 requests per second Python/Tornado = 41,329 requests per second Thanks for sharing btw
  - Cyph0n 11y ago
    Absolutely excellent article. I always keep C++ at the back of my mind in case I need it some day, so I think the list of libraries they used will be useful for me in the future. Thanks.
    - gst 11y ago
      If you want to write fast C++ Web services I recommend a look at Seastar: http://www.seastar-project.org/ http://www.seastar-project.org/
  - jorge_leria 11y ago
    The article is not about serving, but about consuming. Not the same beast.
  - e12e 11y ago
    Interesting article. Not sure that the conclusion is all that solid though: "I quantified that 1 C++ server is roughly equivalent to 40 load-balanced python servers for raw computational power based on our HTTP benchmarking. Thus using C++ can really squeeze all the computational juice out of the underlying hardware to save 1/40 off server costs." Well, for this particular start-up, I think c++ was an excellent choice (especially as they already knew c++!) -- but what if you could still run the service on a single service with python? You would still need 2-3 servers with the c++ version (failover, test etc). Or it might turn out that for production load, you'd need 10 python server-instances. Sure, one server could handle this with c++ -- but you're not actually saving 39/40, you're only saving 9/10, because you didn't need "the full 40". Not to mention, that it seems unlikely that the http request stuff is the limiting factor in a distributed OLAP system. So you might do 100 "OLAPS/s", and have them easily served by a 40Kreq/s python service. "I guess we could have written it in Python to start off with but, economically, it would be a wastage of labor cost and time because, at some stage, we would have to scrap it for a C++ version to get the performance we need. The Python code will have no economic value once scrapped." As mentioned, I think c++ was an excellent choice, but the above assumes they'd have to rewrite the entire system in c++, and scrap all the python. Granted, if you don't know python, but do know c++ very well -- it's doubtful that it would be faster to prototype in python than to just use c++. But if (wild guesstimate) you could get 90% of the features, at 10% the loc/dependencies in python -- who's to say that wouldn't make sense? Maybe parts could be done in c/c++, or the project could move to pypy for enough of a speedup that no complete rewrite would be needed... etc. Overall though, I really think this is a great illustration that high-performance compiled languages offer great performance, and can be rather pleasant to work with.
- merdreubu 11y ago
  Because you can get 540 req/s on Raspberry Pi 2 with Elixir/Phoenix. http://blog.onfido.com/using-cpus-elixir-on-raspberry-pi2/ http://blog.onfido.com/using-cpus-elixir-on-raspberry-pi2/
- coldtea 11y ago
  Because it's like Dr Evil asking the UN leaders for "ONE MILLION DOLLARS" to not destroy the world... https://www.youtube.com/watch?v=cKKHSAE1gIs https://www.youtube.com/watch?v=cKKHSAE1gIs
nathancahill 11y ago
I don't care for PHP as much as the next guy, but it's usually in the top 25 of the web framework benchmark (most of the other top langs are Java, Go and C++): https://www.techempower.com/benchmarks/ https://www.techempower.com/benchmarks/
- 50CNT 11y ago
  Just curious, what's up with this Ur language at both position 1 and 4? Never heard of it, and probably not experienced enough to make sense of it, but how is it that a language that doesn't even have a full official tutorial to its name beat out java, C++ and Go in those rankings by a factor of >2? I'm genuinely curious.
  - nathancahill 11y ago
    Ur (Ur/Web?) seems to be built very specifically for the exact things that this benchmark checks (dynamic web pages with SQL queries). So it's not surprising that those code paths are highly optimized in the language.
kh_hk 11y ago
> Am I missing something? What's so amazing about this? The article is not about testing performance of a web server, but showcasing performance differences between synchronous and asynchronous code using asyncio. So, not about serving requests, but consuming.
- lossolo 11y ago
  Then he should change the title.
  - kh_hk 11y ago
    Making, not serving. I think the title is pretty accurate.

aaossa 11y ago

Why you say is not amazing? Honestly curious here :)

imaginenore 11y ago
Because it's trivial. I would be interested in anything doing 10,000+ req/sec on a cheap VPS. 320 is nothing. People achieve 2 million requests/second with C++ on EC2: https://medium.com/swlh/starting-a-tech-startup-with-c-6b5d5856e6de#.38xs3etwg https://medium.com/swlh/starting-a-tech-startup-with-c-6b5d5...
- aaossa 11y ago
  Oh I see now... This speaks by it self: C++/Proxygen =1,990,130 requests per second Python/Tornado = 41,329 requests per second Thanks for sharing btw
- Cyph0n 11y ago
  Absolutely excellent article. I always keep C++ at the back of my mind in case I need it some day, so I think the list of libraries they used will be useful for me in the future. Thanks.
  - gst 11y ago
    If you want to write fast C++ Web services I recommend a look at Seastar: http://www.seastar-project.org/ http://www.seastar-project.org/
- jorge_leria 11y ago
  The article is not about serving, but about consuming. Not the same beast.
- e12e 11y ago
  Interesting article. Not sure that the conclusion is all that solid though: "I quantified that 1 C++ server is roughly equivalent to 40 load-balanced python servers for raw computational power based on our HTTP benchmarking. Thus using C++ can really squeeze all the computational juice out of the underlying hardware to save 1/40 off server costs." Well, for this particular start-up, I think c++ was an excellent choice (especially as they already knew c++!) -- but what if you could still run the service on a single service with python? You would still need 2-3 servers with the c++ version (failover, test etc). Or it might turn out that for production load, you'd need 10 python server-instances. Sure, one server could handle this with c++ -- but you're not actually saving 39/40, you're only saving 9/10, because you didn't need "the full 40". Not to mention, that it seems unlikely that the http request stuff is the limiting factor in a distributed OLAP system. So you might do 100 "OLAPS/s", and have them easily served by a 40Kreq/s python service. "I guess we could have written it in Python to start off with but, economically, it would be a wastage of labor cost and time because, at some stage, we would have to scrap it for a C++ version to get the performance we need. The Python code will have no economic value once scrapped." As mentioned, I think c++ was an excellent choice, but the above assumes they'd have to rewrite the entire system in c++, and scrap all the python. Granted, if you don't know python, but do know c++ very well -- it's doubtful that it would be faster to prototype in python than to just use c++. But if (wild guesstimate) you could get 90% of the features, at 10% the loc/dependencies in python -- who's to say that wouldn't make sense? Maybe parts could be done in c/c++, or the project could move to pypy for enough of a speedup that no complete rewrite would be needed... etc. Overall though, I really think this is a great illustration that high-performance compiled languages offer great performance, and can be rather pleasant to work with.
merdreubu 11y ago
Because you can get 540 req/s on Raspberry Pi 2 with Elixir/Phoenix. http://blog.onfido.com/using-cpus-elixir-on-raspberry-pi2/ http://blog.onfido.com/using-cpus-elixir-on-raspberry-pi2/
coldtea 11y ago
Because it's like Dr Evil asking the UN leaders for "ONE MILLION DOLLARS" to not destroy the world... https://www.youtube.com/watch?v=cKKHSAE1gIs https://www.youtube.com/watch?v=cKKHSAE1gIs

nathancahill 11y ago

I don't care for PHP as much as the next guy, but it's usually in the top 25 of the web framework benchmark (most of the other top langs are Java, Go and C++): https://www.techempower.com/benchmarks/ https://www.techempower.com/benchmarks/

50CNT 11y ago
Just curious, what's up with this Ur language at both position 1 and 4? Never heard of it, and probably not experienced enough to make sense of it, but how is it that a language that doesn't even have a full official tutorial to its name beat out java, C++ and Go in those rankings by a factor of >2? I'm genuinely curious.
- nathancahill 11y ago
  Ur (Ur/Web?) seems to be built very specifically for the exact things that this benchmark checks (dynamic web pages with SQL queries). So it's not surprising that those code paths are highly optimized in the language.

kh_hk 11y ago

> Am I missing something? What's so amazing about this? The article is not about testing performance of a web server, but showcasing performance differences between synchronous and asynchronous code using asyncio. So, not about serving requests, but consuming.

lossolo 11y ago
Then he should change the title.
- kh_hk 11y ago
  Making, not serving. I think the title is pretty accurate.