netdev - [Netchannels] netchannels scalability.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20061221163338.GB10845@2ka.mipt.ru>
Date:	Thu, 21 Dec 2006 19:33:39 +0300
From:	Evgeniy Polyakov <johnpol@....mipt.ru>
To:	netdev@...r.kernel.org
Cc:	David Miller <davem@...emloft.net>
Subject: [Netchannels] netchannels scalability.

Script has create 22k wildcard (only one dimension (source address) 
is wildcard to save some mem) NAT rules
(16k $rand.$rand.$rand.$rand/255.255.248.0 rules and
(6k $rand.$rand.$rand.$rand/255.255.255.$rand).

I.e. it is equal to 22k following netfilter rules:
iptables -t nat -I POSTROUTING \
	-s $rand.$rand.$rand.$rand/255.255.248.0 \
	-d $static/$some_wildcard_mask \
	--sport $static_num/$some_wildcard_mask \
	--dport $static_num/$some_wildcard_mask \
	--proto $static_num/$some_wildcard_mask \
	-j SNAT --to-source $static_num

$static_num and $some_wildcard_mask are different on each line, but are
the same for different commands.

Raw send/recv performance over gigabit link is the same as described in
previous mail in this thread.

I can not test NAT processing speed itself (I do not have enough test
machines :).

Userspace trie tests included 1-2 millions insertions/searches.

Network performance of the netchannels userspace network stack remains
the same (userspace netchannel is one of the about thousands).

But I managed to crash test system with the latest patchset (netchannel 
addition is not protected against removing as must be done according to
RCU rules).

19 version of the netchannels subsystem patch only contains API
cleanup used with provided test case.

-- 
	Evgeniy Polyakov
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html