<html><head></head><body><div style="font-family: Verdana;font-size: 12.0px;"><div>
<div dir="ltr">
<div>Hello Carlos,</div>
<div> </div>
<div>thx for your reply. I generelly think that gluster should run out of the box with an good speed if the mainsystem has good speed. E.g. if i mesure 300MB/s Gluster should min run with 100MB/s, but your right its a complex thematic.</div>
<div> </div>
<div>As i wrote above i mesure the performance with the Tool "fio" and run some read and write tests. I read about using of XFS, but EXT4 should do it also, while it is normaly as fast as XFS or even faster. Or is glusterfs optimized for XFS? I never worked with XFS, therefore i use EXT4... Yes i mesured first the hardware and than gluster...</div>
<div> </div>
<div>Bevore running Gluster I thought about Hardware and the right Raid. After using a couple of years Raid5 and hat some desaster i decided to use raid6 because of speed and costs. Raid10 is to expensive for me.</div>
<div> </div>
<div>Now i take the 2. Disk from the Controller and put it on en external USB on another system. If i now copy the fils over the network the speed goes up to 30-34MB/s. Could it the Controller?</div>
<div> </div>
<div>------------------------------------------</div>
<div>Actually there are tons of possible tune ups, but let's start from the beginning.</div>
<div>Gluster documents, Red Hat documents, and several other sources suggest using XFS, and NOT ext4.</div>
<div>Second, How did you measure performance ? What tool ?</div>
<div>From my personal experience, first of all, measure performance of you hardware, before measuring Gluster's.</div>
<div>I found out a major problem with my disk layout. I was naive enough believe that by using a good RAID controller (PERC 6) with fair disks (SAS 7.2 KRPM 3 Gbits) on RAID5 I would be in the safe side. It was a disaster. It never passed 40 MBytes/s.</div>
<div>Changing that to RAID10 made it 10 times faster, at the price of half of the storage space. </div>
<div>Test your hardware. </div>
<div>Start with simple stuff like</div>
<div> </div>
<div>dd if=/dev/zero of=filename_on_your_local_disks bs=512k count=10000</div>
<div> </div>
<div>check what the result is, and if your hardware is responding as expected. </div>
<div>Only then move to other areas.</div>
<div>Cheers, </div>
<div> </div>
<div>Carlos</div>
</div>
<br/>
On Tue, Apr 29, 2014 at 10:47 AM, Dragon <span dir="ltr"><<a href="mailto:Sunghost@gmx.de">Sunghost@gmx.de</a>></span> wrote:
<div>
<div style="font-family: Verdana; font-size: 12.0px;">
<div>Hello,</div>
<div> </div>
<div>i have a 3 Node Gluster with one distributed Volume. All 3 Nodes are identically in their Hardware. All have 6x3TB Seagate ST3000DM001. The disk on each Node are in a Softwareraid6 (Blocksize 4096| Stripe 512) with ext4 filesystem on Debian Wheezy. As the Raid was new i testet it with fio:</div>
<div>[sequential-write]<br/>
rw=write<br/>
size=5g<br/>
direct=0<br/>
numjobs=3<br/>
group_reporting<br/>
name=sequential-write-buffered<br/>
bs=4k<br/>
runtime=10</div>
<div> </div>
<div>Result: WRITE: io=3014.2MB, aggrb=308456KB/s, minb=308456KB/s, maxb=308456KB/s,</div>
<div> </div>
<div>I added a new sata controller Digitus DS-30104-1 4 Port PCIe x4, each Port 6GBits on x16 Slot. Node1 has now 7 Disks in Raid6. I added another Disk 3TB same model too on the Controller and copied files 20-40GB onto the volume. The Volume is mounted on the same node with "mount -t gluster-fs IP-Adress:/vol1" The Problem is that the writespeed is only 20MB/s. This is what i didnt understand. The files are copied on the same machine, not over the network, perhaps the bus-system of the mainboard is to slow? Any kind of performance tuning?</div>
<div> </div>
<div>thx</div>
</div>
</div>
</div></div></body></html>