WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-users

RE: [Xen-users] domains gets stalled for N seconds - problem with DMA?

To: <xen-users@xxxxxxxxxxxxxxxxxxx>
Subject: RE: [Xen-users] domains gets stalled for N seconds - problem with DMA?
From: <damjan.pipan.list@xxxxxxx>
Date: Wed, 22 Oct 2008 12:36:44 +0200
Delivery-date: Wed, 22 Oct 2008 03:37:30 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: Ack0MhQdVg/Zxn3pT76XiD+fHfuD3A==
Hi,

I still have problems with io performance. Please tell me if I am doing 
something wrong. It seams that Xen can not run two domains
with a lot of IO usage - a lot of requests not transfer?

Tests:

I have 8 SAS drives and 8 SATA drives. I have created:
- 7 SAS in RAID6 and installed dom0 on this volume group
- 7 SATA in RAID6 and installed domain-1 on this volume group
- 1 SAS as volume and installed domain-2 on this volume group

I have set the scheduler credits
- dom0 - 60000
- dom1 - 1
- dom2 - 1

1 case:
=====
dom1 - write few hundreds of files per minute.
dom2 - idle

IT WORKS NORMALLY - few percent of wait/user in dom1

2 case:
=====
- dom2: tar xfz file.tgz with 3 milion of small files (1 KB)

The domain 1 in this case get from 30-100% wait time. 

IMPORTANT:
- dom1 and dom2 are on different arrays.
- I run also top -d .5 on dom0 ---> it is usually idle but sometimes process 
blkback.17.sda1 gets to the top just for a moment (17
is ID of dom-2)
- I checked also xm top and dom1 is 0 CPU usage when top in dom1 shows 100 wait 
time 

----
Hardware/software is 
2 * Intel(R) Xeon(R) CPU           E5450  @ 3.00GHz
Adaptec 31605 controller

uname -a:
Linux xen-de 2.6.18.8-xen #2 SMP Fri Jun 27 21:27:02 UTC 2008 x86_64 GNU/Linux

host                   : xen-de
release                : 2.6.18.8-xen
version                : #2 SMP Fri Jun 27 21:27:02 UTC 2008
machine                : x86_64
nr_cpus                : 8
nr_nodes               : 1
cores_per_socket       : 4
threads_per_core       : 1
cpu_mhz                : 3000
hw_caps                : 
bfebfbff:20100800:00000000:00000140:000ce3bd:00000000:00000001
total_memory           : 16382
free_memory            : 1028
node_to_cpu            : node0:0-7
xen_major              : 3
xen_minor              : 2
xen_extra              : .1
xen_caps               : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 
hvm-3.0-x86_32p hvm-3.0-x86_64
xen_scheduler          : credit
xen_pagesize           : 4096
platform_params        : virt_start=0xffff800000000000
xen_changeset          : unavailable
cc_compiler            : gcc version 4.1.2 20061115 (prerelease) (Debian 
4.1.1-21)
cc_compile_by          : root
cc_compile_domain      : xlab.si
cc_compile_date        : Fri Jun 27 23:35:01 CEST 2008
xend_config_format     : 4





_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users

<Prev in Thread] Current Thread [Next in Thread>
  • RE: [Xen-users] domains gets stalled for N seconds - problem with DMA?, damjan.pipan.list <=