On 03-09-03 16:31:43 CEST, Michael Bell wrote:
Actually I use one messagequeue and two fifos because I don't know how fast fifos are and how many bytes I can transfer through a messagequeue.
speed of a fifo? about as fast as you can get, i suppose, it's just memory copying.
$ dd if=/dev/zero bs=1k count=1000000 | dd of=/dev/null bs=1k 1000000+0 records in 1000000+0 records in 1000000+0 records out 1024000000 bytes transferred in 3.460183 seconds (295938101 bytes/sec) 1000000+0 records out 1024000000 bytes transferred in 3.461872 seconds (295793719 bytes/sec)
about 300 MB/s, and don't ask me what the bottleneck was. :-)
what do you mean by how many bytes through a message queue? at once? the linux implementation sets the limit to 4080 bytes, looks like one page minus 16 bytes. at all? do you expect to run into problems once you transfer more than 2 or 4 GiB?
Ok, perhaps I should describe what I do.
1. The server uses a messagequeue and waits for incoming messages.
2. If a client needs a value then it sends a fifoname to the messagequeue of the server. The rest of the communication will be handled via two fifo.
The access to the fifos and messagequeue can be restricted very easily via chmod 600 and it looks fast enough. If somebody knows a better way then I'm open for every recommendation. We need a fast local communication between a server and client with the same UID and GID.
Thanks Michael -- ------------------------------------------------------------------- Michael Bell Email: [EMAIL PROTECTED] ZE Computer- und Medienservice Tel.: +49 (0)30-2093 2482 (Computing Centre) Fax: +49 (0)30-2093 2704 Humboldt-University of Berlin Unter den Linden 6 10099 Berlin Email (private): [EMAIL PROTECTED] Germany http://www.openca.org
------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ OpenCA-Devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/openca-devel