<div dir="ltr"><div>Nope, not a caching thing. I've tried multiple different types of fio tests, all produce the same results. Gbps when hitting the disks locally, slow MB\s when hitting the Gluster FUSE mount.</div><div><br></div><div>I've been reading up on glustr-ganesha, and will give that a try.<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Dec 12, 2023 at 1:58 PM Ramon Selga <<a href="mailto:ramon.selga@gmail.com">ramon.selga@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><u></u>
<div>
<font face="Futura Bk BT">Dismiss my first question: you have SAS
12Gbps SSDs Sorry!</font><br>
<br>
<div>El 12/12/23 a les 19:52, Ramon Selga ha
escrit:<br>
</div>
<blockquote type="cite">
<font face="Futura Bk BT">May ask you which kind of disks you have
in this setup? rotational, ssd SAS/SATA, nvme?<br>
<br>
Is there a RAID controller with writeback caching?<br>
<br>
It seems to me your fio test on local brick has a unclear result
due to some caching.<br>
<br>
Try something like (you can consider to increase test file size
depending of your caching memory) :<br>
<br>
fio --size=16G --name=test --filename=/gluster/data/brick/wow
--bs=1M --nrfiles=1 --direct=1 --sync=0 --randrepeat=0
--rw=write --refill_buffers --end_fsync=1 --iodepth=200
--ioengine=libaio<br>
<br>
</font>Also remember a replica 3 arbiter 1 volume writes
synchronously to two data bricks, halving throughput of your
network backend.<br>
<br>
Try similar fio on gluster mount but I hardly see more than
300MB/s writing sequentially on only one fuse mount even with nvme
backend. On the other side, with 4 to 6 clients, you can easily
reach 1.5GB/s of aggregate throughput <br>
<br>
To start, I think is better to try with default parameters for
your replica volume.<br>
<br>
Best regards!<br>
<br>
Ramon<br>
<br>
<br>
<div>El 12/12/23 a les 19:10, Danny ha
escrit:<br>
</div>
<blockquote type="cite">
<div dir="ltr">Sorry, I noticed that too after I posted, so I
instantly upgraded to 10. Issue remains. <br>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Tue, Dec 12, 2023 at
1:09 PM Gilberto Ferreira <<a href="mailto:gilberto.nunes32@gmail.com" target="_blank">gilberto.nunes32@gmail.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div dir="ltr">I strongly suggest you update to version 10
or higher. <br>
It's come with significant improvement
regarding performance.<br clear="all">
<div>
<div dir="ltr" class="gmail_signature">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div>---</div>
<div>
<div>
<div>Gilberto Nunes Ferreira</div>
</div>
<div><span style="font-size:12.8px">(47)
99676-7530 - Whatsapp / Telegram</span><br>
</div>
<div>
<p style="font-size:12.8px;margin:0px"><br>
</p>
<p style="font-size:12.8px;margin:0px"><br>
</p>
</div>
</div>
<div><br>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<br>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">Em ter., 12 de dez. de
2023 às 13:03, Danny <<a href="mailto:dbray925%2Bgluster@gmail.com" target="_blank">dbray925+gluster@gmail.com</a>>
escreveu:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div dir="ltr"> MTU is already 9000, and as you can see
from the IPERF results, I've got a nice, fast
connection between the nodes. </div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Tue, Dec 12, 2023
at 9:49 AM Strahil Nikolov <<a href="mailto:hunter86_bg@yahoo.com" target="_blank">hunter86_bg@yahoo.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div> Hi,
<div><br>
</div>
<div>Let’s try the simple things:</div>
<div><br>
</div>
<div>Check if you can use MTU9000 and if it’s
possible, set it on the Bond Slaves and the bond
devices:</div>
<div><span> ping GLUSTER_PEER </span><span>-c 10
-M do -s 8972</span></div>
<div><br>
Then try to follow up the recommendations from <a href="https://access.redhat.com/documentation/en-us/red_hat_gluster_storage/3.5/html/administration_guide/chap-configuring_red_hat_storage_for_enhancing_performance" target="_blank">https://access.redhat.com/documentation/en-us/red_hat_gluster_storage/3.5/html/administration_guide/chap-configuring_red_hat_storage_for_enhancing_performance</a>
<div><br>
</div>
<div><br>
</div>
Best Regards,</div>
<div>Strahil Nikolov <br>
<br>
<p style="font-size:15px;color:rgb(113,95,250);padding-top:15px;margin-top:0px">On
Monday, December 11, 2023, 3:32 PM, Danny <<a href="mailto:dbray925%2Bgluster@gmail.com" target="_blank">dbray925+gluster@gmail.com</a>>
wrote:</p>
<blockquote>
<div id="m_-4149661103284099357m_-3283052153129726045m_98609290250261231m_810164770777586547yiv1869372156">
<div dir="ltr">
<div>Hello list, I'm hoping someone can
let me know what setting I missed.</div>
<div><br>
</div>
<div>Hardware:</div>
<div>Dell R650 servers, Dual 24 Core Xeon
2.8 GHz, 1 TB RAM<br>
</div>
<div>8x SSD s <span>Negotiated Speed</span>
12 Gbps</div>
<div>PERC H755 Controller - RAID 6 <br>
</div>
<div>Created virtual "data" disk from the
above 8 SSD drives, for a ~20 TB
/dev/sdb<br>
</div>
<div><br>
</div>
<div>OS:</div>
<div>CentOS Stream</div>
<div>kernel-4.18.0-526.el8.x86_64</div>
<div>glusterfs-7.9-1.el8.x86_64</div>
<div><br>
</div>
<div>IPERF Test between nodes:<br>
[ ID] Interval Transfer
Bitrate Retr<br>
[ 5] 0.00-10.00 sec 11.5 GBytes
9.90 Gbits/sec 0 sender<br>
[ 5] 0.00-10.04 sec 11.5 GBytes
9.86 Gbits/sec
receiver<br>
</div>
<div><br>
</div>
<div>All good there. ~10 Gbps, as
expected.<br>
</div>
<div><br>
</div>
<div>LVM Install:</div>
<div>export DISK="/dev/sdb"<br>
sudo parted --script $DISK "mklabel gpt"<br>
sudo parted --script $DISK "mkpart
primary 0% 100%"<br>
sudo parted --script $DISK "set 1 lvm
on"</div>
<div>sudo pvcreate --dataalignment 128K
/dev/sdb1<br>
sudo vgcreate --physicalextentsize 128K
gfs_vg /dev/sdb1<br>
sudo lvcreate -L 16G -n gfs_pool_meta
gfs_vg<br>
sudo lvcreate -l 95%FREE -n gfs_pool
gfs_vg<br>
sudo lvconvert --chunksize 1280K
--thinpool gfs_vg/gfs_pool
--poolmetadata gfs_vg/gfs_pool_meta<br>
sudo lvchange --zero n gfs_vg/gfs_pool<br>
sudo lvcreate -V 19.5TiB --thinpool
gfs_vg/gfs_pool -n gfs_lv<br>
sudo mkfs.xfs -f -i size=512 -n
size=8192 -d su=128k,sw=10
/dev/mapper/gfs_vg-gfs_lv<br>
sudo vim /etc/fstab</div>
<div>/dev/mapper/gfs_vg-gfs_lv
/gluster/data/brick xfs
rw,inode64,noatime,nouuid 0 0</div>
<div><br>
</div>
<div>sudo systemctl daemon-reload
&& sudo mount -a<br>
fio --name=test
--filename=/gluster/data/brick/wow
--size=1G --readwrite=write<br>
</div>
<div><br>
</div>
<div>Run status group 0 (all jobs):<br>
WRITE: bw=2081MiB/s (2182MB/s),
2081MiB/s-2081MiB/s (2182MB/s-2182MB/s),
io=1024MiB (1074MB), run=492-492msec<br>
</div>
<div><br>
</div>
<div>All good there. 2182MB/s =~ 17.5
Gbps. Nice!<br>
</div>
<div><br>
</div>
<div><br>
</div>
<div>Gluster install:</div>
<div>export NODE1='10.54.95.123'<br>
export NODE2='10.54.95.124'<br>
export NODE3='10.54.95.125'<br>
sudo gluster peer probe $NODE2<br>
sudo gluster peer probe $NODE3<br>
sudo gluster volume create data replica
3 arbiter 1 $NODE1:/gluster/data/brick
$NODE2:/gluster/data/brick
$NODE3:/gluster/data/brick force<br>
sudo gluster volume set data
network.ping-timeout 5<br>
sudo gluster volume set data
performance.client-io-threads on<br>
sudo gluster volume set data group
metadata-cache<br>
sudo gluster volume start data<br>
sudo gluster volume info all<br>
</div>
<div><br>
Volume Name: data<br>
Type: Replicate<br>
Volume ID:
b52b5212-82c8-4b1a-8db3-52468bc0226e<br>
Status: Started<br>
Snapshot Count: 0<br>
Number of Bricks: 1 x (2 + 1) = 3<br>
Transport-type: tcp<br>
Bricks:<br>
Brick1: 10.54.95.123:/gluster/data/brick<br>
Brick2: 10.54.95.124:/gluster/data/brick<br>
Brick3: 10.54.95.125:/gluster/data/brick
(arbiter)<br>
Options Reconfigured:<br>
network.inode-lru-limit: 200000<br>
performance.md-cache-timeout: 600<br>
performance.cache-invalidation: on<br>
performance.stat-prefetch: on<br>
features.cache-invalidation-timeout: 600<br>
features.cache-invalidation: on<br>
network.ping-timeout: 5<br>
transport.address-family: inet<br>
storage.fips-mode-rchecksum: on<br>
nfs.disable: on<br>
performance.client-io-threads: on</div>
<div><br>
</div>
<div>sudo vim /etc/fstab<br>
</div>
<div>localhost:/data /data
glusterfs defaults,_netdev
0 0</div>
<div><br>
</div>
<div>sudo systemctl daemon-reload
&& sudo mount -a</div>
<div>fio --name=test --filename=/data/wow
--size=1G --readwrite=write</div>
<div><br>
</div>
<div>Run status group 0 (all jobs):<br>
WRITE: bw=109MiB/s (115MB/s),
109MiB/s-109MiB/s (115MB/s-115MB/s),
io=1024MiB (1074MB), run=9366-9366msec</div>
<div><br>
</div>
<div>Oh no, what's wrong? From 2182MB/s
down to only 115MB/s? What am I missing?
I'm not expecting the above ~17 Gbps,
but I'm thinking it should at least be
close(r) to ~10 Gbps. <br>
</div>
<div><br>
</div>
<div>Any suggestions?</div>
</div>
</div>
________<br>
<br>
<br>
<br>
Community Meeting Calendar:<br>
<br>
Schedule -<br>
Every 2nd and 4th Tuesday at 14:30 IST / 09:00
UTC<br>
Bridge: <a href="https://meet.google.com/cpu-eiue-hvk" target="_blank">https://meet.google.com/cpu-eiue-hvk</a><br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users" target="_blank">https://lists.gluster.org/mailman/listinfo/gluster-users</a><br>
</blockquote>
</div>
</div>
</blockquote>
</div>
________<br>
<br>
<br>
<br>
Community Meeting Calendar:<br>
<br>
Schedule -<br>
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC<br>
Bridge: <a href="https://meet.google.com/cpu-eiue-hvk" rel="noreferrer" target="_blank">https://meet.google.com/cpu-eiue-hvk</a><br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">https://lists.gluster.org/mailman/listinfo/gluster-users</a><br>
</blockquote>
</div>
</blockquote>
</div>
<br>
<fieldset></fieldset>
<pre>________
Community Meeting Calendar:
Schedule -
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC
Bridge: <a href="https://meet.google.com/cpu-eiue-hvk" target="_blank">https://meet.google.com/cpu-eiue-hvk</a>
Gluster-users mailing list
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users" target="_blank">https://lists.gluster.org/mailman/listinfo/gluster-users</a>
</pre>
</blockquote>
<br>
</blockquote>
<br>
</div>
________<br>
<br>
<br>
<br>
Community Meeting Calendar:<br>
<br>
Schedule -<br>
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC<br>
Bridge: <a href="https://meet.google.com/cpu-eiue-hvk" rel="noreferrer" target="_blank">https://meet.google.com/cpu-eiue-hvk</a><br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">https://lists.gluster.org/mailman/listinfo/gluster-users</a><br>
</blockquote></div>