My understanding is that the block layer in Linux does not allow block sizes larger then the page size of the system. Therefore on Intel and zSeries hardware 4096 is the largest possible block size. I think this is the same for pSeries as well.
If you could have a larger block size it would probably not be worth it anyway since the average wasted space for files on a filesystem is one half of the block size times the number of files. In other words: you waste one half a block for each file, on average. If you have a larger block you waste more space for each file. -----Original Message----- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Nix, Robert P. Sent: Wednesday, August 10, 2005 12:34 PM To: LINUX-390@VM.MARIST.EDU Subject: Dasdfmt and other potential block sizes I've been letting the blocksize default to 4096 for dasdfmt, but I'm getting abuse from zOS types for not using half-track blocking, and the space I'm losing to IRGs. What are the possible values for the blocksize, and what are the pros and cons of changing away from the default of 4096? Anyone doing this? -- Robert P. Nix Mayo Foundation RO-CE-8-857 200 First Street SW 507-284-0844 Rochester, MN 55905 ----- "In theory, theory and practice are the same, but in practice, theory and practice are different." ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
