From owner-freebsd-scsi Thu Oct 22 09:08:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA14980 for freebsd-scsi-outgoing; Thu, 22 Oct 1998 09:08:51 -0700 (PDT) (envelope-from owner-freebsd-scsi@FreeBSD.ORG) Received: from cse.psu.edu (claven.cse.psu.edu [130.203.3.50]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA14972 for ; Thu, 22 Oct 1998 09:08:47 -0700 (PDT) (envelope-from keefe@cse.psu.edu) Received: from remulak.cse.psu.edu (keefe@remulak.cse.psu.edu [130.203.30.18]) by cse.psu.edu (8.8.8/8.8.8) with ESMTP id MAA27281; Thu, 22 Oct 1998 12:08:03 -0400 (EDT) From: Thomas F Keefe Received: (from keefe@localhost) by remulak.cse.psu.edu (8.8.8/8.8.7) id MAA05083; Thu, 22 Oct 1998 12:08:02 -0400 (EDT) Date: Thu, 22 Oct 1998 12:08:02 -0400 (EDT) Message-Id: <199810221608.MAA05083@remulak.cse.psu.edu> To: patton@sysnet.net Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: Sequential Disk I/O Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > if you study how the big boys liek Oracle do it, they use block sizes in > excess of 8kb. Depending on how big your disks are 32kb blocks are > considered more reasonable. What they do is fit multiple database "blocks" > or records into a logical block. I suggest a vastly bigger blocksize than > 512bytes if you're serious about performance. By using large blocks I can approximate the performance of sequentail access. I may have a seek and rotation penalty at the beginning of the transfer, but that one penalty is amortized over a large number of sectors. This may be my only choice if I cannot solve this problem. At this point I have spent a lot of time and become very curious about this. What I was hoping to get from this group, was something like: (1) It is impossible to avoid the rotational latency when issuing writes to adjacent sectors on a SCSI disk because the time required between the completion of one command and the start of the next is a significant fraction of the time it takes for a 5400RPM disk platter to rotate. Thus, even with large strides, the next command comes to late. This, makes only one access per revolution possible. or; (2) Modern SCSI drives (by default) access an entire track on both reads and writes. Thus, only one access is possible per revolution. This default behavior can be disabled through the mode page as follows ... or; (3) I have done this, the trick is to ... If anyone could rule out (1) or (2) above, that would also help a lot. Any enlightenment you can offer on this topic will be appreciated. Tom Keefe Dept. of Computer Science and Engineering Penn State keefe@cse.psu.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message