atom feed45 messages in org.freebsd.freebsd-fsRe: Areca vs. ZFS performance testing.
FromSent OnAttachments
Jeremy ChadwickOct 30, 2008 8:31 pm 
Danny CarrollOct 30, 2008 9:07 pm 
Jeremy ChadwickOct 30, 2008 9:33 pm 
Andrew SnowOct 30, 2008 9:43 pm 
Danny CarrollOct 30, 2008 9:47 pm 
Danny CarrollOct 30, 2008 9:49 pm 
Danny CarrollOct 30, 2008 9:54 pm 
Simun MikecinOct 31, 2008 2:20 am 
Simun MikecinOct 31, 2008 4:56 am 
Peter SchullerNov 2, 2008 7:08 am 
Simun MikecinNov 3, 2008 12:31 am 
DieterNov 12, 2008 2:57 pm 
Danny CarrollNov 12, 2008 9:46 pm 
Jeremy ChadwickNov 12, 2008 11:42 pm 
Willem Jan WithagenNov 13, 2008 12:32 am 
Danny CarrollNov 13, 2008 3:09 am 
Danny CarrollNov 13, 2008 5:58 am 
Nikolay DenevNov 13, 2008 7:05 am 
Scott LongNov 13, 2008 8:49 am 
Danny CarrollNov 13, 2008 12:46 pm 
Danny CarrollNov 13, 2008 12:59 pm 
Eirik ØverbyNov 16, 2008 12:26 pm 
Danny CarrollNov 16, 2008 7:15 pm 
Matt SimersonNov 16, 2008 10:06 pm 
Jeremy ChadwickNov 16, 2008 11:07 pm 
Wes MorganNov 17, 2008 3:26 am 
Danny CarrollNov 17, 2008 3:42 am 
Matt SimersonNov 17, 2008 1:04 pm 
Matt SimersonNov 17, 2008 2:07 pm 
Danny CarrollNov 17, 2008 3:45 pm 
Jan MikkelsenDec 2, 2008 2:38 am 
Wes MorganDec 2, 2008 4:04 am 
Danny CarrollJan 7, 2009 4:33 pm 
Zaphod BeeblebroxJan 7, 2009 11:40 pm 
Koen SmitsJan 7, 2009 11:48 pm 
Nikolay DenevJan 8, 2009 1:19 am 
Danny CarrollJan 8, 2009 6:29 pm 
Koen SmitsJan 9, 2009 12:46 am 
Danny CarrollJan 9, 2009 1:02 am 
Koen SmitsJan 9, 2009 7:57 am 
Andrew SnowJan 9, 2009 6:38 pm 
Danny CarrollJan 9, 2009 8:58 pm 
Danny CarrollJan 20, 2009 10:40 pm 
Koen SmitsJan 21, 2009 1:15 am 
Danny CarrollJan 21, 2009 5:14 am 
Subject:Re: Areca vs. ZFS performance testing.
From:Danny Carroll (fb@dannysplace.net)
Date:Nov 13, 2008 12:59:10 pm
List:org.freebsd.freebsd-fs

Scott Long wrote:

The Areca controller likely doesn't buffer/cache for disks in JBOD mode, as others in this thread have stated. Without buffering, simple disk controllers will almost always be faster than accelerated raid controllers because the accelerated controllers add more latency between the host and the disk. A simple controller will directly funnel data from the host to the disk as soon as it receives a command. An accelerated controller, however, has a CPU and a mini-OS on it that has to schedule the work coming from the host and handle its own tasks and interrupts. This adds latency that quickly adds up under benchmarks. Your numbers clearly demonstrate this.

That's nice to know. I'm not sure it tells us why the Non-Cached writes were about 8% faster though. The other thing about the "NoWriteCache" test I performed that I neglected to mention yesterday is that I actually panic'd the box (running out of memory). This was the first time I have had that happen with ZFS even though in previous testing (with cache enabled) I punished the box for a lot longer.

Perhaps the ZFS caching took over where the disk caching left off? Could that explain why I did not see a negative difference in the numbers between Cache enabled and Cache disabled?

One of the questions I wanted to answer for myself was just this: "Does a battery-backed cache on an Areca card protect me when I am in JBOD mode." If the Areca does not buffer/cache in JBOD mode then that means the answer is no.

-D