Tag Info

Hot answers tagged

5

If you're using nginx, then you're talking just a few KB of overhead per active connection. If you're using something like Apache, you'll have one thread per connection, which means hundreds of KB or even megabytes per connection. However, nginx does not support asynchronous disk IO on Linux (because async disk IO on Linux is basically horribly broken by ...


3

There's a G-WAN rewrite example here (look at the second handler source code, the first handler example illustrates FLV pseudo-streaming).


2

If the OS can cache the hot part of the content into ram, it will not use the disk and will serve things really quickly. Hundreds of request per second should be possible on a VPS, you will most likely saturate the network well before you run into CPU limits. If the content does not fit into ram, then disk IO (seek, throughput, filesystem fragmentation) ...


2

This comparative test (G-WAN vs Nginx) was done by an Academic expert of distributde architectures. This is the very same test used by Phoronix, the official multi-OS benchmark.


2

This link states that G-WAN can scale better than nginx and it can benefit more from multi-core CPUs and uses less resources. You can have a look at the architecture and performance sections. In my opinion, I recommend you to install anyone of them and do your own benchmarks on your estimated/expected loads.


1

I am not sure but I believe that the www. prefix is hard-coded as an alias already. You should rather try with something like secure.movq.org (providing that your DNS records are set, and that the client will use this domain name as the "Host: xxx" HTTP header then it should work for you too). To simplify things you can use the G-WAN aliases feature for ...


1

The reason for this to happen comes from the fact that your virtualized environment has 287.62 MiB RAM free. This old release of G-WAN was not tuned for hypervisors (virtualization) and this leaded to an infinite loop in the memory manager (which assumed AMD64 servers would have more than 1 GB RAM). Yo resolve your problem you can either allocate 1 GB RAM ...


1

OK, I've found the solution after more research. All I have to do is adding my main.c file into handlers directory of the hostname folder. Thanks to those guys, who have been asked and answered to the followings: http://stackoverflow.com/questions/14847994/custom-404-response-for-gwan ...


1

using an [AMD] Opteron 6234 6 module / 12 core processor This score for the 6-Core AMD Opteron @ 2.4GHz[1] 137,913 req/s falls short of our 850,000 req/s on an Intel 6-Core Xeon W3680 @ 3.33GHz[2] (with a 100-byte static file). Besides the differences of each architure performance*, the problem for G-WAN comes on AMD CPUs from the fact that we did not ...


1

The discussions you may have seen are correct... See here, here, here, and here. You can't change glibc versions on CentOS 5. You'll need an upgrade to a newer version of RHEL/CentOS. EL6 ships with glibc 2.12: Name : glibc Relocations: (not relocatable) Version : 2.12 Vendor: CentOS Of ...


1

You should try to remove the www. prefix. To avoid the need for trivial rewrites, G-WAN translates www.domain requests into domain requests so there's no need to specify www. explicitely. If you REALLY need to use distinct sites for www.domain and domain then use a different IP address and G-WAN listener (G-WAN can listen to as many IP addresses as you ...


1

My 2 cents: set up and test with local wgets or other scripted http requesting that either discard the downloaded files or dump them on a RAM disk (make sure you do not make the OS page out the RAM disk). Experience values: on modern hardware, nginx will saturate a Gigabit link EASILY with multi-MB files. So could aio-era lighttpd, which is discontinued and ...


1

You can Use a virtual host, or even an alias, see the G-WAN FAQs. For more elaborate rewrites, you can use a handler, some examples are provided in the download archive.


1

what resources are needed for serving static content? How much RAM is used for each file transfer? First, for the same number of workers, G-WAN v4.7+ is using far less RAM than Nginx at startup: > Server 'nginx' process topology: --------------------------------------------- 6] pid:21228 Process RAM: 0.77 MB 5] pid:21229 Process RAM: 2.44 MB ...



Only top voted, non community-wiki answers of a minimum length are eligible