Re: RSS controller v2 Test results (lmbench )

From: Balbir Singh <balbir@linux.vnet.ibm.com>
To: Rik van Riel <riel@redhat.com>
Cc: Pavel Emelianov <xemul@sw.ru>, Paul Menage <menage@google.com>,
	Kirill Korotaev <dev@sw.ru>,
	devel@openvz.org, Linux Containers <containers@lists.osdl.org>,
	linux kernel mailing list <linux-kernel@vger.kernel.org>,
	Linux Memory Management List <linux-mm@kvack.org>,
	Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Herbert Poetzl <herbert@13thfloor.at>
Subject: Re: RSS controller v2 Test results (lmbench )
Date: Fri, 18 May 2007 09:37:22 +0530	[thread overview]
Message-ID: <464D267A.50107@linux.vnet.ibm.com> (raw)
In-Reply-To: <464D1599.1000506@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 1522 bytes --]

Rik van Riel wrote:
> Balbir Singh wrote:
> 
>> A meaningful container size does not hamper performance. I am in the
>> process
>> of getting more results (with varying container sizes). Please let me
>> know
>> what you think of the results? Would you like to see different
>> benchmarks/
>> tests/configuration results?
> 
> AIM7 results might be interesting, especially when run to crossover.
> 

I'll try and get hold of AIM7, I have some AIM9 results (please
see the attachment, since the results overflow 80 columns, I've
attached them).

> OTOH, AIM7 can make the current VM explode spectacularly :)
> 
> I saw it swap out 1.4GB of memory in one run, on my 2GB memory test
> system.  That's right, it swapped out almost 75% of memory.
> 

This would make a good test case for the RSS and the unmapped page
cache controller. Thanks for bringing it to my attention.

> Presumably all the AIM7 processes got stuck in the pageout code
> simultaneously and all decided they needed to swap some pages out.
> However, the shell got stuck too so I could not get sysrq output
> on time.
> 

oops! I wonder if AIM7 creates too many processes and exhausts all
memory. I've seen a case where during an upgrade of my tetex on my
laptop, the setup process failed and continued to fork processes
filling up 4GB of swap.

> I am trying out a little VM patch to fix that now, carefully watching
> vmstat output.  Should be fun...
> 

VM debugging is always fun!

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL

[-- Attachment #2: aim9 --]
[-- Type: text/plain, Size: 3365 bytes --]

------------------------------------------------------------------------------------------------------------
 Test        Test        Elapsed  Iteration    Iteration          Operation
Number       Name      Time (sec)   Count   Rate (loops/sec)    Rate (ops/sec)
------------------------------------------------------------------------------------------------------------
     1 creat-clo           60.00       8885  148.08333       148083.33 File Creations and Closes/second  (256 MB container)
     1 creat-clo           60.01       8547  142.42626       142426.26 File Creations and Closes/second  (unlimited container)
     1 creat-clo           60.01       8632  143.84269       143842.69 File Creations and Closes/second  (container not mounted)
     2 page_test           60.00       6068  101.13333       171926.67 System Allocations & Pages/second (256 MB container)
     2 page_test           60.00       5275   87.91667       149458.33 System Allocations & Pages/second  (unlimited container)
     2 page_test           60.01       5411   90.16831       153286.12 System Allocations & Pages/second  (container not mounted)
     3 brk_test            60.01       9151  152.49125      2592351.27 System Memory Allocations/second  (256 MB container)
     3 brk_test            60.02       7404  123.35888      2097100.97 System Memory Allocations/second  (unlimited container)
     3 brk_test            60.01       8294  138.21030      2349575.07 System Memory Allocations/second  (container not mounted)
     4 jmp_test            60.01     983062 16381.63639     16381636.39 Non-local gotos/second  (256 MB container)
     4 jmp_test            60.00     983084 16384.73333     16384733.33 Non-local gotos/second  (unlimited container)
     4 jmp_test            60.00     982904 16381.73333     16381733.33 Non-local gotos/second  (container not mounted)
     5 signal_test         60.01      28013  466.80553       466805.53 Signal Traps/second  (256 MB container)
     5 signal_test         60.00      28360  472.66667       472666.67 Signal Traps/second  (unlimited container)
     5 signal_test         60.01      28593  476.47059       476470.59 Signal Traps/second  (container not mounted)
     6 exec_test           60.02       2596   43.25225          216.26 Program Loads/second  (256 MB container)
     6 exec_test           60.02       2539   42.30257          211.51 Program Loads/second  (unlimited container)
     6 exec_test           60.01       2536   42.25962          211.30 Program Loads/second  (container not mounted)
     7 fork_test           60.01       2118   35.29412         3529.41 Task Creations/second  (256 MB container)
     7 fork_test           60.03       2130   35.48226         3548.23 Task Creations/second  (unlimited container)
     7 fork_test           60.01       2130   35.49408         3549.41 Task Creations/second  (container not mounted)
     8 link_test           60.02      47760  795.73476        50131.29 Link/Unlink Pairs/second  (256 MB container)
     8 link_test           60.02      48156  802.33256        50546.95 Link/Unlink Pairs/second  (unlimited container)
     8 link_test           60.00      49778  829.63333        52266.90 Link/Unlink Pairs/second  (container not mounted)
------------------------------------------------------------------------------------------------------------