Re: [Libevent-users] http: libevent vs many threads
On Fri, Apr 18, 2008 at 10:36:53PM +0200, Cezary Rzewuski wrote: > Thank you for Your suggestions. I've just finished the implementation. > I used the approach of libevent as HTTP server and threads working > on downloaded content (they are performing some statistical computation on > downloaded javascripts). It looks to work efficiently. > > It's probably not the right group, but you says that switching between > threads > is expensive. However, I've read somewhere (it was probably "Advanced linux > programming" by Alex Samuel) that creating a new thread is nearly as fast as > calling a function. Does it mean, that switching between threads is > slower than > creating a new thread? Creating a new thread is almost certainly not as fast as calling a function. Maybe, I suppose, as fast calling into the kernel; author's point being that Linux can quickly allocate and setup the task structures. But this sounds like one of those things you should test yourself. I'm not quite sure what the benchmark would look like. Try Usenet: comp.unix.programmer, or comp.programming.threads. ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users
Re: [Libevent-users] http: libevent vs many threads
On Sat, Apr 19, 2008 at 02:49:20AM +0200, Springande Ulv wrote: > > On 19. april. 2008, at 01.57, William Ahern wrote: > >In some sense--like code complexity--using both an event-oriented > >design > >*and* threads is the worst of both worlds. Your processing logic is > >turned > >inside-out (state machines, etc, can be confusing to some people), > >plus you > >have mutexes and barriers and all that crap littered throughout your > >code. > > He he, the crack you just heard is the thin ice you walked out on. I must be deaf. > >Premature optimization is the root of all evil > > Oh that is an old misunderstood meme. What Hoare actually said was "We > should forget about small efficiencies, say about 97% of the time: > premature optimization is the root of all evil." He was talking about > micro-optimizing code. Not exactly what I would call thinking about a > scalable program design, which one should think about from the start > and at all time, unless you work in the Microsoft Office team. I really couldn't care less what Hoare or anybody else said. I'm not appealing to authority. It's an apposite phrase. If you fancy all your first or even second iterations of your projects running a gigantic data center, *and* you actually meet your targets out the gate, then good for you. (Normal people would write a proof-of-concept first, and maybe second, and third.) But, in my experience it's best to aim for correctness first. If you must, you can fix most any well designed single process event-oriented daemon to scale [more] by tweaking it to use multiple processes, or multiple servers. The same more-or-less goes for threaded servers. But you usually don't need to, because as long as you're not needlessly copying data or doing other wasteful things, then you'll best any second-rate application, period. ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users
Re: [Libevent-users] http: libevent vs many threads
On 19. april. 2008, at 01.57, William Ahern wrote: In some sense--like code complexity--using both an event-oriented design *and* threads is the worst of both worlds. Your processing logic is turned inside-out (state machines, etc, can be confusing to some people), plus you have mutexes and barriers and all that crap littered throughout your code. He he, the crack you just heard is the thin ice you walked out on. Premature optimization is the root of all evil Oh that is an old misunderstood meme. What Hoare actually said was "We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil." He was talking about micro-optimizing code. Not exactly what I would call thinking about a scalable program design, which one should think about from the start and at all time, unless you work in the Microsoft Office team. ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users
Re: [Libevent-users] http: libevent vs many threads
On Sat, Apr 19, 2008 at 12:23:16AM +0200, Springande Ulv wrote: > > On 18. april. 2008, at 22.36, Cezary Rzewuski wrote: > >>our best bet is to use a single thread using > >>libevent, or go totally multi-threaded without libevent. In 90% of > >>the > >>circumstances one of those options (though not both) > > Why not both?! Using both threads _and_ libevent is not only possible > both often feasible. Use libevent in a i/o bound thread and a thread > pool for CPU bound operations. In a network application the network is > the bottleneck and a few context switches is diminutive in this context. Because _unless_ you can justify the additional complexity, why bother? In some sense--like code complexity--using both an event-oriented design *and* threads is the worst of both worlds. Your processing logic is turned inside-out (state machines, etc, can be confusing to some people), plus you have mutexes and barriers and all that crap littered throughout your code. I've used libevent with threads. I've also used libevent with multi-process configurations. But, usually I stick with one or the other. Premature optimization is the root of all evil ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users
Re: [Libevent-users] http: libevent vs many threads
On 18. april. 2008, at 22.36, Cezary Rzewuski wrote: our best bet is to use a single thread using libevent, or go totally multi-threaded without libevent. In 90% of the circumstances one of those options (though not both) Why not both?! Using both threads _and_ libevent is not only possible both often feasible. Use libevent in a i/o bound thread and a thread pool for CPU bound operations. In a network application the network is the bottleneck and a few context switches is diminutive in this context. ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users
Re: [Libevent-users] http: libevent vs many threads
Thank you for Your suggestions. I've just finished the implementation. I used the approach of libevent as HTTP server and threads working on downloaded content (they are performing some statistical computation on downloaded javascripts). It looks to work efficiently. It's probably not the right group, but you says that switching between threads is expensive. However, I've read somewhere (it was probably "Advanced linux programming" by Alex Samuel) that creating a new thread is nearly as fast as calling a function. Does it mean, that switching between threads is slower than creating a new thread? Once more - thanks for comprehensive answer. William Ahern wrote: On Wed, Apr 02, 2008 at 10:15:59PM +0200, Cezary Rzewuski wrote: Hi, I'd like to ask if sending http requests with libevent is carried out in separate thread or is the library single-threaded? I want to use the library in a program which will visit many URL and download it's content. Is it good idea to use libevent or the classic solution with creating a separate thread per URL request will be much more efficient solution? It depends. What you describe is not nearly enough informatio to even give a suggestion. One thread per URL normally is a very poor choice (just as a matter of runtime efficiency), unless each URL causes you to do a lot of disk I/O, or if each URL causes you to do CPU intensive operations, like decode compressed audio/video. In each of those two situations, the process context switching costs are diminished relative to the type of work being done. Basically, the idea is that if your thread will block on an operation--CPU or I/O--but another thread running in parallel (not merely concurrently) could utilize additional resources, you want to multi-thread. If your application is merely moving bytes (say, as a proxy), usually a single thread is enough; you can multiplex non-blocking network operations on a single thread. In that sense, you're "switching contexts" in the application, and not the kernel. This reduces the workload, because context switching in the kernel is usually more expensive., OTOH, copying data in itself can be CPU intensive. If you read into a buffer from one socket, you might evict previous data you read in earlier. If you then try to re-read and/or copy that previous data over to another buffer later, the process will block as the data is fetched from RAM. If your proxy is even on a 100Mb connection, depending on how you process the data, you most definitely will need multiple threads. That's because 100Mb of network data could ballon to 5x or 10x that mount of byte shuffling. Of course, depending on how the L1, L2 and L3 caches are shared, it might not actually make much of a difference. It all depends! Of course, you can always use an event-oriented model within each particular thread. Or spread event delivery and processing across multiple threads. Given that you seem new to this (or at least new to the particular problem you're trying to solve), your best bet is to use a single thread using libevent, or go totally multi-threaded without libevent. In 90% of the circumstances one of those options (though not both) are as near to optimal as you'll get, and you don't need to the headaches of any additional complexity. I saw that libevent was used in spybye, which is kind of similar what I want to do. I was wondering if spybye were more efficient with requests served in separate threads instead of using libevent (I don't say that it's not efficient, just theoretically). I'm not sure, maybe its most efficient using _both_. But I suspect it probably just uses libevent in a single thread. Note, there are other ways to use threads. You could use one thread using libevent to handle all your queries and network I/O. Then you could use a separate thread worker pool to, for instance, run ClamAV on the data. This works well if you can isolate your CPU intensive work outside the mundane network I/O parts. If your application is overall CPU bound, and latency of particular requests isn't of primary concern, then it doesn't matter that libevent is running in a single thread. All your CPUs are doing work, just not the same types of work. ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users ___ Libevent-users mailing list Libevent-users@monkey.org http://monkeymail.org/mailman/listinfo/libevent-users