RE: I/O and db_file_multiblock_read_count

From: Matthew Zito <mzito_at_gridapp.com>
Date: Tue, 12 Dec 2006 12:39:36 -0500
Message-ID: <C0A5E31718FC064A91E9FD7BE2F081B199CABC@exchange.gridapp.com>

/dev/null is hard, as some OS'es, solaris notably, do evil tricks with certain utilities (like cat something to /dev/null - it returns immediately). Not sure about dd

---

        From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Kevin Closson

	Sent: Tuesday, December 12, 2006 12:35 PM
	To: oracle-l
	Subject: RE: I/O and db_file_multiblock_read_count
	
	
	I would have never though a write to /dev/zero would work...I
learn something new every day.
	Can you try this test writing to /dev/null?


________________________________

		From: Jesse, Rich [mailto:Rich.Jesse_at_qg.com] 
		Sent: Tuesday, December 12, 2006 7:16 AM
		To: kevinc_at_polyserve.com; oracle-l
		Subject: RE: I/O and db_file_multiblock_read_count
		
		
		I'm guessing that I'm limited by CPU on this IBM JS21

blade's LPAR (MPV, 2 cores max) with an SVC (virtualized SAN) backend. topas showed kernel mode CPU >95% for most of the tests.                  

/oracle $ time dd if=S.dbf of=/dev/zero bs=1024k

		16384+1 records in
		16384+1 records out
		 
		real    1m3.044s
		user    0m0.271s
		sys     0m42.458s
		


		/oracle $ time dd if=S.dbf of=/dev/zero bs=1024k

		16384+1 records in
		

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

Received on Tue Dec 12 2006 - 11:39:36 CST