RE: Sun T2000

From: Kevin Closson <kevinc_at_polyserve.com>
Date: Sun, 25 Mar 2007 09:52:39 -0700
Message-ID: <5D2570CAFC98974F9B6A759D1C74BAD002F7449A@ex2.ms.polyserve.com>

Interesting that the fetch count is different. Since there was no PIO, the Proliant would likely get better performance without indirect data buffers BTW.

I'd like to see somehting that runs a bit longer such as JL Computing Index
http://www.miraclebenelux.nl/jloci.html I'd also like to have measurements from my silly little benchmark (http://kevinclosson.wordpress.com/2007/01/30/oracle-on-opteron-with-lin ux-the-numa-angle-part-iii/)

any chance for that?

-----Original Message-----
From: Alexander Fatkulin [mailto:afatkulin_at_gmail.com] Sent: Saturday, March 24, 2007 9:02 PM
To: Kevin Closson
Cc: oracle-l_at_freelists.org
Subject: Re: Sun T2000

Kevin,

thanks for the question. There were no PIO involved.

Here is what I see:

create table t as

	select mod(level, 10) m10,
			mod(level, 25) m25,
			mod(level, 50) m50,
			level l
		from dual
		connect by level <= 500000;

Proliant results:

select m10, m25, m50, sum(l), count(*), min(l), max(l), avg(l)  from t
 group by m10, m25, m50

call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ----------

---

Parse 1 0.00 0.00 0 1 0 0
Execute 1 0.00 0.00 0 0 0 0
Fetch 5 0.69 0.67 0 649 0 50
------- ------ -------- ---------- ---------- ---------- ----------

---

total 7 0.69 0.67 0 650 0 50

T2000 results:

select m10, m25, m50, sum(l), count(*), min(l), max(l), avg(l)  from t
 group by m10, m25, m50

call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ----------

---

Parse 1 0.00 0.00 0 0 0 0
Execute 1 0.00 0.00 0 0 0 0
Fetch 2 4.45 4.34 0 649 0 50
------- ------ -------- ---------- ---------- ---------- ----------

---

total 4 4.45 4.34 0 649 0 50

Both DB were created with a 16K blocksize. And mind that - Proliant were running with use_indirect_data_buffers=true.

On 3/25/07, Kevin Closson <kevinc_at_polyserve.com> wrote:
> of the T2000 server
> vs Proliant DL380 G3 (based on 3Ghz Xeons).
>
> One Niagara core were approximately 7 to 8 times slower than single
> Xeon CPU in batch task (huge sorts/aggregates) -
>
> ...I'm not syaing you didn't see this result, but that seem slower
than
> expected. To what degree was I/O a factor?
>
>
>
>
>
>

-- 
Alexander Fatkulin
--
http://www.freelists.org/webpage/oracle-l

Received on Sun Mar 25 2007 - 11:52:39 CDT