Re: Raw Disk still preferred for 11.5?
On 6 Oct, 20:56, "Neil Truby" <neil.tr...@ardenta.com> wrote: > "InDeep" <ind...@indeep.com> wrote in message > > news:48ea5464$0$5171$6c36adad@news.usenetserver.com... > > > If you are on a SAN, raw disk is really not necessary, it may yield a > > small performance gain, > > Hmmm. I though most of the perceived gain in performance waas supposed to > come from bypassing the operating system's I/O operations. In which case, > why would it make any difference what the underlying storage system is? > > To address the original question, I have it in mind that the orthodoxy these > days was that the File System cacheing that was thought to slow down cooked > speeds was now avoidable in modern versions of Linux (such as the one you > propose) anyway and there was little difference betweeen one and the other. > > Usually I read things like " ...I have been seeing and hearing only around > 10% improvement [therefore] I'd actually recommend against raw disks" but > never really understand them, as even if there was minimal performance > benefit I'd still probably keep things raw because: > a. There's less likeliehood that someone will try to access the files > directly and > b. It makes things a little more arcane and therefore makes it slightly more > difficult to take the support away from me (especially now Informix is like > DB2 and completely autonomic and so doesn't need any administration ;-)) > > Fernando Nunes seems to know what he's talking about:http://informix-technology.blogspot.com/2007/05/cheetah-spot-by-spot-... Informix http://publib.boulder.ibm.com/infocenter/idshelp/v115/topic/com.ibm.admin.doc/ids_admin_0480.htm?resultof=%22%72%61%77%22%20%22%64%69%73%6b%22%20 "While it is recommended that you use raw disk devices on UNIX to achieve better performance..." DB2 http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/topic/com.ibm.db2.luw.admin.config.doc/doc/r0000246.html?resultof=%22%72%61%77%22%20%22%64%69%73%6b%22%20 "Note that as of DB2® Version 9, the use of raw devices for database logging is deprecated" So how come using raw disk for logs is recommended for Informix (and for better performance) but for DB2 is deprecated when both products come from the same company...answer me that IBM!! So if raw devices are faster for Informix and DB2 no longer supports them for database logging then isn't Informix faster than DB2 when you do a lot of logged operations i.e. high volume OLTP??? Answer me that as well IBM... Ha! Ha! Ha!
david@smooth1.co.uk wrote: > On 6 Oct, 20:56, "Neil Truby" <neil.tr...@ardenta.com> wrote: >> "InDeep" <ind...@indeep.com> wrote in message >> >> news:48ea5464$0$5171$6c36adad@news.usenetserver.com... >> >>> If you are on a SAN, raw disk is really not necessary, it may yield a >>> small performance gain, >> Hmmm. I though most of the perceived gain in performance waas supposed to >> come from bypassing the operating system's I/O operations. In which case, >> why would it make any difference what the underlying storage system is? >> >> To address the original question, I have it in mind that the orthodoxy these >> days was that the File System cacheing that was thought to slow down cooked >> speeds was now avoidable in modern versions of Linux (such as the one you >> propose) anyway and there was little difference betweeen one and the other. >> >> Usually I read things like " ...I have been seeing and hearing only around >> 10% improvement [therefore] I'd actually recommend against raw disks" but >> never really understand them, as even if there was minimal performance >> benefit I'd still probably keep things raw because: >> a. There's less likeliehood that someone will try to access the files >> directly and >> b. It makes things a little more arcane and therefore makes it slightly more >> difficult to take the support away from me (especially now Informix is like >> DB2 and completely autonomic and so doesn't need any administration ;-)) >> >> Fernando Nunes seems to know what he's talking about:http://informix-technology.blogspot.com/2007/05/cheetah-spot-by-spot-... > > Informix > http://publib.boulder.ibm.com/infocenter/idshelp/v115/topic/com.ibm.admin.doc/ids_admin_0480.htm?resultof=%22%72%61%77%22%20%22%64%69%73%6b%22%20 > > "While it is recommended that you use raw disk devices on UNIX to > achieve better performance..." > > DB2 > http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/topic/com.ibm.db2.luw.admin.config.doc/doc/r0000246.html?resultof=%22%72%61%77%22%20%22%64%69%73%6b%22%20 > > "Note that as of DB2' Version 9, the use of raw devices for database > logging is deprecated" > > So how come using raw disk for logs is recommended for Informix (and > for better performance) but for DB2 is deprecated when both products > come from the same company...answer me that IBM!! > > So if raw devices are faster for Informix and DB2 no longer supports > them for database logging then isn't Informix faster than DB2 > when you do a lot of logged operations i.e. high volume OLTP??? Answer > me that as well IBM... > > Ha! Ha! Ha! We could laugh about this, but if we take the time to read a bit more your DB2 link we see: "The string can point to either a path name or to a raw device. Note that as of DB2' Version 9, the use of raw devices for database logging is deprecated. As an alternative to using raw logs, you can use either direct input/output (DIO) or concurrent input/output (CIO)." And the Informix: "Important: While it is recommended that you use raw disk devices on UNIX to achieve better performance, if you enable the DIRECT_IO configuration parameter, the performance for cooked files can approach the performance of raw devices used for dbspace chunks. This occurs because direct I/O bypasses the use of the file system buffers. For more information, see the IBM' Informix' Dynamic Server Performance Guide." So, taking into account that the use of DIRECT_IO tries to force DIO and CIO, the only difference is that the Informix guys haven't yet "deprecated" the raw devices... More than laughing about it, I'd be curious to know why... If you care to look for one or two AIX papers about RAW/DIO/CIO, you may find it very interesting... I believe in a not too distant future nobody will be using RAW... But let's wait and see. Regards, -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently...
Hello Fernando, does DIRECT_IO bypasses the filesystem cache??? using cooked can kill performance because linux/unix is being hit (to) hard. one example is do a restore on char mode raw dev and cooked specially on Sun. Cooked restore is a factor 2/3 or more slower!!! besides this the machine is suffering when the restore is done. On linux this is no different and one can not limit the amount of memory used for fs cache... well not that i am aware of like hp does with DBC_MAX_PCT and DBC_MIN_PCT. Also on aix i could do 50 mb/sec writes to char mode raw (yeah yeah 7 years ago...) while the aix teacher did 40 mb/sec writing to the filesystem using dd.....and that was really the best he could do.... So if filesystem caches etc are not seriously improved since the last 7 years i still would gofor char mode raw always. and not being lazy Superboer. On 8 okt, 01:35, Fernando Nunes <domusonl...@gmail.com> wrote: > da...@smooth1.co.uk wrote: > > On 6 Oct, 20:56, "Neil Truby" <neil.tr...@ardenta.com> wrote: > >> "InDeep" <ind...@indeep.com> wrote in message > > >>news:48ea5464$0$5171$6c36adad@news.usenetserver.com... > > >>> If you are on a SAN, raw disk is really not necessary, it may yield a > >>> small performance gain, > >> Hmmm. I though most of the perceived gain in performance waas supposed to > >> come from bypassing the operating system's I/O operations. In which case, > >> why would it make any difference what the underlying storage system is? > > >> To address the original question, I have it in mind that the orthodoxy these > >> days was that the File System cacheing that was thought to slow down cooked > >> speeds was now avoidable in modern versions of Linux (such as the one you > >> propose) anyway and there was little difference betweeen one and the other. > > >> Usually I read things like " ...I have been seeing and hearing only around > >> 10% improvement [therefore] I'd actually recommend against raw disks" but > >> never really understand them, as even if there was minimal performance > >> benefit I'd still probably keep things raw because: > >> a. There's less likeliehood that someone will try to access the files > >> directly and > >> b. It makes things a little more arcane and therefore makes it slightly more > >> difficult to take the support away from me (especially now Informix is like > >> DB2 and completely autonomic and so doesn't need any administration ;-)) > > >> Fernando Nunes seems to know what he's talking about:http://informix-technology.blogspot.com/2007/05/cheetah-spot-by-spot-... > > > Informix > >http://publib.boulder.ibm.com/infocenter/idshelp/v115/topic/com.ibm.a... > > > "While it is recommended that you use raw disk devices on UNIX to > > achieve better performance..." > > > DB2 > >http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/topic/com.ibm.db... > > > "Note that as of DB2® Version 9, the use of raw devices for database > > logging is deprecated" > > > So how come using raw disk for logs is recommended for Informix (and > > for better performance) but for DB2 is deprecated when both products > > come from the same company...answer me that IBM!! > > > So if raw devices are faster for Informix and DB2 no longer supports > > them for database logging then isn't Informix faster than DB2 > > when you do a lot of logged operations i.e. high volume OLTP??? Answer > > me that as well IBM... > > > Ha! Ha! Ha! > > We could laugh about this, but if we take the time to read a bit more your DB2 > link we see: > > "The string can point to either a path name or to a raw device. Note that as of > DB2® Version 9, the use of raw devices for database logging is deprecated. As > an alternative to using raw logs, you can use either direct input/output (DIO) > or concurrent input/output (CIO)." > > And the Informix: > > "Important: While it is recommended that you use raw disk devices on UNIX to > achieve better performance, if you enable the DIRECT_IO configuration > parameter, the performance for cooked files can approach the performance of raw > devices used for dbspace chunks. This occurs because direct I/O bypasses the > use of the file system buffers. For more information, see the IBM® Informix® > Dynamic Server Performance Guide." > > So, taking into account that the use of DIRECT_IO tries to force DIO and CIO, > the only difference is that the Informix guys haven't yet "deprecated" the raw > devices... More than laughing about it, I'd be curious to know why... If you > care to look for one or two AIX papers about RAW/DIO/CIO, you may find it very > interesting... > > I believe in a not too distant future nobody will be using RAW... But let's > wait and see. > Regards, > > -- > Fernando Nunes > Portugal > > http://informix-technology.blogspot.com > My email works... but I don't check it frequently...