Let's Play a Game

testing the s3700

The Setup

Two 100G s3700 drives, one tested with luks, one not.

Filled the drive to test with it filled.

Testing 4k/8k, with luks using --size=8 or --size=9 for 4k and 8k respectivly.

I used the following settings in fio, changing the filename and block size where appropriate.

[global]
bs=4k
ioengine=posixaio
iodepth=32
size=200g
filename=/dev/mapper/testssd
direct=1

[rand-read]
rw=randread
stonewall

[rand-write]
rw=randwrite
stonewall

[seq-read]
rw=read
stonewall

[seq-write]
rw=write
stonewall

Results

There was generally high interupts and context switches, but oddly less so with luks.

Sequential IOPS

4k sequential writes with luks was 73.7% of max ( 12424 vs 16855 )
4k sequential reads with luks was 76.7% of max ( 14471 vs 18864 )
8k sequential writes with luks was 71.0% of max ( 9640 vs 13573 )
8k sequential reads with luks was 71.8% of max ( 10744 vs 14966 )

Random IOPS

4k random writes with luks was 82.2% of max ( 13919 vs 16924
4k random reads with luks was 80.7% of max ( 6260 vs 7756 )
8k random writes with luks was 71.7% of max ( 9718 vs 13557 )
8k random reads with luks was 64.7% of max ( 4222 vs 6526 )

Conclusion

My use case is zfs usage as l2arc/zil cache, I'll be using 8k on luks.

Other articles

Interupts and Context Switches with ZFS with LUKS
Published: 2014 February 20 Thu
By Matthew Thode (prometheanfire)
In General.
tags: zfs
Why care?
Interupts and Context switches represent overhead and ineffeciency (in general).
The Setup
The Same 15 4TB drives in a raidz3 with a hotspare (16 drives total).
Testing was just a scrub (no fresh boot). Switched between CFQ and NOOP a couple of times.
Results
The picture generally speaks …
read more
ZFS performance on LUKS
Published: 2014 February 18 Tue
By Matthew Thode (prometheanfire)
In General.
tags: zfs
the setup
The system being tested is a 15 drive raidz3 setup with 4TB drives running kernel 3.13.2 (non-hardened for testing).
The encryption algorithm
I was previously using cbc-essiv, this causes I found (via a simple scrub test) that it caused my performance to be about 10% lower …
read more
Well, here we go again.
Published: 2012 July 20 Fri
By Matthew Thode (prometheanfire)
In General.
Well, here we go again, I think (hope) to actually keep this blog more up to date with the little projects I do. Going to be writing about selinux zfsonlinux the 'cloud' and whatever strikes my fancy.
I supose I should introduce myself, name's Matthew Thode, I currently work for …
read more

Other articles

Why care?

The Setup

Results

the setup

The encryption algorithm

social