How to calculate the buffer wait ratio?

Posted in 2000

Summary

A newcomer asked how to derive the buffer-wait ratio from 'onstat -p'. Replies reposted an awk script (by Allen Jantzen, formulas credited to Art Kagel) that parses onstat -p output and prints two figures: Bufwaits Ratio = bufwaits/(pagreads+bufwrits)*100, which should ideally stay under about 7%, and read-ahead utilization, which should be near 100%. A follow-up correction established that the RA formula in the original script was inverted; the correct form is RA-pgsused/(ixda-RA+idx-RA+da-RA)*100, and Kagel confirmed it — a result over 100% simply means the calculation was done backwards.

Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.

Topics: General Discussion

JasonYLPang@pg.SLR.com — Wed, 26 Jul 2000 11:32:06 +0800 — source: Informix-list mailing list archive (1991-1998)

This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01BFF6B2.12B03494 Content-Type: text/plain; charset="iso-8859-1" Hi all, I'm a newbie here...can anyone out there teach me how to calculate the bufwaits ratio based on the output of onstat -p? Informix Dynamic Server Version 7.30.UC5 -- On-Line -- Up 2 days 18:20:25 -- 1686080 Kbytes Profile dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached 337544121 22939961 1286467833 73.76 28977960 20018155 356806201 91.88 isamtot open start read write rewrite delete commit rollbk 2962803697 50018416 637440143 380886011 160748285 90253778 13290496 26197003 1045675 gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs 0 0 0 0 0 0 0 ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes 0 0 0 1023616.34 54671.99 784 1568 bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans 66639943 1282 618021591 24 0 9252 1324794 12935294 ixda-RA idx-RA da-RA RA-pgsused lchwaits 168610454 13181361 80669720 261997622 144597997 Thanks/Regards, Jason ------_=_NextPart_001_01BFF6B2.12B03494 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Diso-8859-1"> <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = 5.5.2650.12"> <TITLE>How to calculate the buffer wait ratio?</TITLE> </HEAD> <BODY> Hi all, I'm a newbie here...can anyone out = there teach me how to calculate the bufwaits ratio based on the output = of onstat -p? Informix Dynamic Server Version = 7.30.UC5   -- On-Line -- Up 2 days 18:20:25 -- 1686080 = Kbytes Profile dskreads pagreads bufreads %cached = dskwrits pagwrits bufwrits %cached 337544121 22939961 1286467833 = 73.76   28977960 20018155 356806201 91.88  isamtot  = open     start    = read     write    rewrite  = delete   commit   rollbk 2962803697 50018416 637440143 = 380886011 160748285 90253778 13290496 26197003 1045675 gp_read  gp_write gp_rewrt = gp_del   gp_alloc gp_free  gp_curs 0        = 0        = 0        = 0        = 0        = 0        = 0       ovlock   ovuserthread = ovbuff   usercpu  syscpu   numckpts flushes = 0        = 0            = 0        1023616.34 54671.99 = 784      1568    bufwaits lokwaits lockreqs = deadlks  dltouts  ckpwaits compress seqscans 66639943 = 1282     618021591 = 24       = 0        = 9252     1324794  12935294 ixda-RA  idx-RA   = da-RA    RA-pgsused lchwaits 168610454 13181361 80669720 = 261997622  144597997 Thanks/Regards, Jason </BODY> </HTML> ------_=_NextPart_001_01BFF6B2.12B03494--

↪ replying to JasonYLPang@pg.SLR.com

Marat — Wed, 26 Jul 2000 13:49:34 +0400 — source: Informix-list mailing list archive (1991-1998)

It's from previous posting. It's very helpful for me. And I won't tire to say my thanks to the authors. Ok - a contribution to all the helpful CDI folks. The following Awk script will show you the UR (read-ahead utilization ratio) and the BR (bufwaits ratio). Art Kagel gets all the credit for the formulas - I just got sick of using my calculator. ---------------------------------------- awk ' /[a-zA-Z]/ { for(i=1; i<=NF; i++){ name[i] = $i; } } /[0-9]/ { for(i=1; i<=NF; i++){ content[name[i]] = $i; } } END { ixdaRA = content["ixda-RA"]; idxRA = content["idx-RA"]; daRA = content["da-RA"]; RApgsused = content["RA-pgsused"]; print "Read Utilization (UR): ", ((ixdaRA + idxRA + daRA) / RApgsused) * 100, "%"; bufwaits = content["bufwaits"]; bufwrits = content["bufwrits"]; pagreads = content["pagreads"]; print "Bufwaits Ratio(BR): ", ((bufwaits/(pagreads + bufwrits)) * 100), "%"; } ' ---------------------------------------------------------- The UR should ideally be VERY near 100% - the higher the better. The BR should be below 7% - the lower the better. Check CDI archives for more info. JasonYLPang@pg.SLR.com wrote: > This message is in MIME format. Since your mail reader does not understand > this format, some or all of this message may not be legible. > > ------_=_NextPart_001_01BFF6B2.12B03494 > Content-Type: text/plain; > charset="iso-8859-1" > > Hi all, > > I'm a newbie here...can anyone out there teach me how to calculate the > bufwaits ratio based on the output of onstat -p? > > Informix Dynamic Server Version 7.30.UC5 -- On-Line -- Up 2 days 18:20:25 > -- 1686080 Kbytes > > Profile > dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached > 337544121 22939961 1286467833 73.76 28977960 20018155 356806201 91.88 > > isamtot open start read write rewrite delete commit > rollbk > 2962803697 50018416 637440143 380886011 160748285 90253778 13290496 26197003 > 1045675 > > gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs > 0 0 0 0 0 0 0 > > ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes > 0 0 0 1023616.34 54671.99 784 1568 > > bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans > 66639943 1282 618021591 24 0 9252 1324794 12935294 > > ixda-RA idx-RA da-RA RA-pgsused lchwaits > 168610454 13181361 80669720 261997622 144597997 > > Thanks/Regards, > Jason > > ------_=_NextPart_001_01BFF6B2.12B03494 > Content-Type: text/html; > charset="iso-8859-1" > Content-Transfer-Encoding: quoted-printable > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> > <HTML> > <HEAD> > <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = > charset=3Diso-8859-1"> > <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = > 5.5.2650.12"> > <TITLE>How to calculate the buffer wait ratio?</TITLE> > </HEAD> > <BODY> > > Hi all, > > > I'm a newbie here...can anyone out = > there teach me how to calculate the bufwaits ratio based on the output = > of onstat -p? > > > Informix Dynamic Server Version = > 7.30.UC5   -- On-Line -- Up 2 days 18:20:25 -- 1686080 = > Kbytes > > > Profile > dskreads pagreads bufreads %cached = > dskwrits pagwrits bufwrits %cached > 337544121 22939961 1286467833 = > 73.76   28977960 20018155 356806201 91.88  > > > isamtot  = > open     start    = > read     write    rewrite  = > delete   commit   rollbk > 2962803697 50018416 637440143 = > 380886011 160748285 90253778 13290496 26197003 1045675 > > > gp_read  gp_write gp_rewrt = > gp_del   gp_alloc gp_free  gp_curs > FACE=3D"Courier">0        = > 0        = > 0        = > 0        = > 0        = > 0        = > 0       > > > ovlock   ovuserthread = > ovbuff   usercpu  syscpu   numckpts flushes = > > FACE=3D"Courier">0        = > 0            = > 0        1023616.34 54671.99 = > 784      1568    > > > bufwaits lokwaits lockreqs = > deadlks  dltouts  ckpwaits compress seqscans > 66639943 = > 1282     618021591 = > 24       = > 0        = > 9252     1324794  12935294 > > > ixda-RA  idx-RA   = > da-RA    RA-pgsused lchwaits > 168610454 13181361 80669720 = > 261997622  144597997 > > > Thanks/Regards, > Jason > > > </BODY> > </HTML> > ------_=_NextPart_001_01BFF6B2.12B03494--

↪ replying to Marat

Art S. Kagel — Wed, 26 Jul 2000 14:40:42 -0400 — source: Informix-list mailing list archive (1991-1998)

Great script, well crafted. I especially love the perl style associative array. I would make three suggestions: 1) add the 'onstat -p | ' to the beginning of the script 2) include the comments about acceptable values in the END{} printout, I did when I installed your script in production. 3) Sign it, add a README or some comments at the top, and submit it officially to the IIUG Repository by emailing it to: software@iiug.org and Walt will install it in the Repository. Art S. Kagel Marat wrote: > > It's from previous posting. It's very helpful for me. And I won't tire to say > my thanks to the authors. > > Ok - a contribution to all the helpful CDI folks. The following Awk > script will show you the UR (read-ahead utilization ratio) and the BR > (bufwaits ratio). Art Kagel gets all the credit for the formulas - I > just got sick of using my calculator. > ---------------------------------------- > awk ' > /[a-zA-Z]/ { > for(i=1; i<=NF; i++){ > name[i] = $i; > } > } > /[0-9]/ { > for(i=1; i<=NF; i++){ > content[name[i]] = $i; > } > } > END { > ixdaRA = content["ixda-RA"]; > idxRA = content["idx-RA"]; > daRA = content["da-RA"]; > RApgsused = content["RA-pgsused"]; > print "Read Utilization (UR): ", ((ixdaRA + idxRA + daRA) / > RApgsused) * 100, "%"; > > bufwaits = content["bufwaits"]; > bufwrits = content["bufwrits"]; > pagreads = content["pagreads"]; > print "Bufwaits Ratio(BR): ", ((bufwaits/(pagreads + bufwrits)) * > 100), "%"; > } > ' > ---------------------------------------------------------- > The UR should ideally be VERY near 100% - the higher the better. > The BR should be below 7% - the lower the better. > Check CDI archives for more info. > > JasonYLPang@pg.SLR.com wrote: > > > This message is in MIME format. Since your mail reader does not understand > > this format, some or all of this message may not be legible. > > > > ------_=_NextPart_001_01BFF6B2.12B03494 > > Content-Type: text/plain; > > charset="iso-8859-1" > > > > Hi all, > > > > I'm a newbie here...can anyone out there teach me how to calculate the > > bufwaits ratio based on the output of onstat -p? > > > > Informix Dynamic Server Version 7.30.UC5 -- On-Line -- Up 2 days 18:20:25 > > -- 1686080 Kbytes > > > > Profile > > dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached > > 337544121 22939961 1286467833 73.76 28977960 20018155 356806201 91.88 > > > > isamtot open start read write rewrite delete commit > > rollbk > > 2962803697 50018416 637440143 380886011 160748285 90253778 13290496 26197003 > > 1045675 > > > > gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs > > 0 0 0 0 0 0 0 > > > > ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes > > 0 0 0 1023616.34 54671.99 784 1568 > > > > bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans > > 66639943 1282 618021591 24 0 9252 1324794 12935294 > > > > ixda-RA idx-RA da-RA RA-pgsused lchwaits > > 168610454 13181361 80669720 261997622 144597997 > > > > Thanks/Regards, > > Jason > > > > ------_=_NextPart_001_01BFF6B2.12B03494 > > Content-Type: text/html; > > charset="iso-8859-1" > > Content-Transfer-Encoding: quoted-printable > > > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> > > <HTML> > > <HEAD> > > <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = > > charset=3Diso-8859-1"> > > <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = > > 5.5.2650.12"> > > <TITLE>How to calculate the buffer wait ratio?</TITLE> > > </HEAD> > > <BODY> > > > > Hi all, > > > > > > I'm a newbie here...can anyone out = > > there teach me how to calculate the bufwaits ratio based on the output = > > of onstat -p? > > > > > > Informix Dynamic Server Version = > > 7.30.UC5   -- On-Line -- Up 2 days 18:20:25 -- 1686080 = > > Kbytes > > > > > > Profile > > dskreads pagreads bufreads %cached = > > dskwrits pagwrits bufwrits %cached > > 337544121 22939961 1286467833 = > > 73.76   28977960 20018155 356806201 91.88  > > > > > > isamtot  = > > open     start    = > > read     write    rewrite  = > > delete   commit   rollbk > > 2962803697 50018416 637440143 = > > 380886011 160748285 90253778 13290496 26197003 1045675 > > > > > > gp_read  gp_write gp_rewrt = > > gp_del   gp_alloc gp_free  gp_curs > > > FACE=3D"Courier">0        = > > 0        = > > 0        = > > 0        = > > 0        = > > 0        = > > 0       > > > > > > ovlock   ovuserthread = > > ovbuff   usercpu  syscpu   numckpts flushes = > > > > > FACE=3D"Courier">0        = > > 0            = > > 0        1023616.34 54671.99 = > > 784      1568    > > > > > > bufwaits lokwaits lockreqs = > > deadlks  dltouts  ckpwaits compress seqscans > > 66639943 = > > 1282     618021591 = > > 24       = > > 0        = > > 9252     1324794  12935294 > > > > > > ixda-RA  idx-RA   = > > da-RA    RA-pgsused lchwaits > > 168610454 13181361 80669720 = > > 261997622  144597997 > > > > > > Thanks/Regards, >

↪ replying to Art S. Kagel

allenj — Thu, 27 Jul 2000 13:23:13 GMT — source: Informix-list mailing list archive (1991-1998)

Actually, I wrote the script and posted it on Nov 20 1999... You can see my original post at http://www.iiug.org/members/cdi_archive/1999.11/cdii.46414 Allen Jantzen, DBA Ned Davis Research "Art S. Kagel" wrote: > > Great script, well crafted. I especially love the perl style > associative array. I would make three suggestions: > > 1) add the 'onstat -p | ' to the beginning of the script > 2) include the comments about acceptable values in the END{} printout, > I did when I installed your script in production. > 3) Sign it, add a README or some comments at the top, and submit it > officially to the IIUG Repository by emailing it to: > software@iiug.org and Walt will install it in the Repository. > > Art S. Kagel > > Marat wrote: > > > > It's from previous posting. It's very helpful for me. And I won't tire to say > > my thanks to the authors. > > > > Ok - a contribution to all the helpful CDI folks. The following Awk > > script will show you the UR (read-ahead utilization ratio) and the BR > > (bufwaits ratio). Art Kagel gets all the credit for the formulas - I > > just got sick of using my calculator. > > ---------------------------------------- > > awk ' > > /[a-zA-Z]/ { > > for(i=1; i<=NF; i++){ > > name[i] = $i; > > } > > } > > /[0-9]/ { > > for(i=1; i<=NF; i++){ > > content[name[i]] = $i; > > } > > } > > END { > > ixdaRA = content["ixda-RA"]; > > idxRA = content["idx-RA"]; > > daRA = content["da-RA"]; > > RApgsused = content["RA-pgsused"]; > > print "Read Utilization (UR): ", ((ixdaRA + idxRA + daRA) / > > RApgsused) * 100, "%"; > > > > bufwaits = content["bufwaits"]; > > bufwrits = content["bufwrits"]; > > pagreads = content["pagreads"]; > > print "Bufwaits Ratio(BR): ", ((bufwaits/(pagreads + bufwrits)) * > > 100), "%"; > > } > > ' > > ---------------------------------------------------------- > > The UR should ideally be VERY near 100% - the higher the better. > > The BR should be below 7% - the lower the better. > > Check CDI archives for more info. > > > > JasonYLPang@pg.SLR.com wrote: > > > > > This message is in MIME format. Since your mail reader does not understand > > > this format, some or all of this message may not be legible. > > > > > > ------_=_NextPart_001_01BFF6B2.12B03494 > > > Content-Type: text/plain; > > > charset="iso-8859-1" > > > > > > Hi all, > > > > > > I'm a newbie here...can anyone out there teach me how to calculate the > > > bufwaits ratio based on the output of onstat -p? > > > > > > Informix Dynamic Server Version 7.30.UC5 -- On-Line -- Up 2 days 18:20:25 > > > -- 1686080 Kbytes > > > > > > Profile > > > dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached > > > 337544121 22939961 1286467833 73.76 28977960 20018155 356806201 91.88 > > > > > > isamtot open start read write rewrite delete commit > > > rollbk > > > 2962803697 50018416 637440143 380886011 160748285 90253778 13290496 26197003 > > > 1045675 > > > > > > gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs > > > 0 0 0 0 0 0 0 > > > > > > ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes > > > 0 0 0 1023616.34 54671.99 784 1568 > > > > > > bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans > > > 66639943 1282 618021591 24 0 9252 1324794 12935294 > > > > > > ixda-RA idx-RA da-RA RA-pgsused lchwaits > > > 168610454 13181361 80669720 261997622 144597997 > > > > > > Thanks/Regards, > > > Jason > > > > > > ------_=_NextPart_001_01BFF6B2.12B03494 > > > Content-Type: text/html; > > > charset="iso-8859-1" > > > Content-Transfer-Encoding: quoted-printable > > > > > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> > > > <HTML> > > > <HEAD> > > > <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = > > > charset=3Diso-8859-1"> > > > <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = > > > 5.5.2650.12"> > > > <TITLE>How to calculate the buffer wait ratio?</TITLE> > > > </HEAD> > > > <BODY> > > > > > > Hi all, > > > > > > > > > I'm a newbie here...can anyone out = > > > there teach me how to calculate the bufwaits ratio based on the output = > > > of onstat -p? > > > > > > > > > Informix Dynamic Server Version = > > > 7.30.UC5   -- On-Line -- Up 2 days 18:20:25 -- 1686080 = > > > Kbytes > > > > > > > > > Profile > > > dskreads pagreads bufreads %cached = > > > dskwrits pagwrits bufwrits %cached > > > 337544121 22939961 1286467833 = > > > 73.76   28977960 20018155 356806201 91.88  > > > > > > > > > isamtot  = > > > open     start    = > > > read     write    rewrite  = > > > delete   commit   rollbk > > > 2962803697 50018416 637440143 = > > > 380886011 160748285 90253778 13290496 26197003 1045675 > > > > > > > > > gp_read  gp_write gp_rewrt = > > > gp_del   gp_alloc gp_free  gp_curs > > > > > FACE=3D"Courier">0        = > > > 0        = > > > 0        = > > > 0        = > > > 0        = > > > 0        = > > > 0       > > > > > > > > > ovlock   ovuserthread = > > > ovbuff   usercpu  syscpu   numckpts flushes = > > > > > > > > FACE=3D"Courier">0        = > > > 0            = > > > 0        1023616.34 54671.99 = > > > 784      1568    > > > > > > > > > bufwaits lokwaits lockreqs = > > > deadlks  dltouts  ckpwaits compress seqscans > > > 66639943 = > > > 1282&nbs

↪ replying to allenj

Zandy — Thu, 27 Jul 2000 08:24:09 -0700 — source: Informix-list mailing list archive (1991-1998)

I'm not sure about this but isn't the formula for RA utilization equal to: (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 and not ((ixdaRA + idxRA + daRA) / RApgsused) * 100 as written in the script? Zandy ----------------------------------------------------------- Got questions? Get answers over the phone at Keen.com. Up to 100 minutes free! http://www.keen.com

↪ replying to Zandy

allenj — Thu, 27 Jul 2000 15:40:21 GMT — source: Informix-list mailing list archive (1991-1998)

I was wondering about this myself, because I was consistently getting a little more than 100% RA utilization. (100.63%) However, using your version of the formula, I get 99.36%. So I asked Art Kagel to confirm the formula that I used, and he said it was OK. We're probably splitting hairs here, but I wonder which one is correct? Art.....any opinions? Allen Zandy wrote: > > I'm not sure about this but isn't the formula for RA utilization > equal to: > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > and not > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > as written in the script? > > Zandy > > ----------------------------------------------------------- > > Got questions? Get answers over the phone at Keen.com. > Up to 100 minutes free! > http://www.keen.com

↪ replying to allenj

Art S. Kagel — Thu, 27 Jul 2000 12:03:50 -0400 — source: Informix-list mailing list archive (1991-1998)

allenj wrote: > > I was wondering about this myself, because I was consistently getting > a little more than 100% RA utilization. (100.63%) > > However, using your version of the formula, I get 99.36%. > > So I asked Art Kagel to confirm the formula that I used, and he said it > was OK. > > We're probably splitting hairs here, but I wonder which one is correct? > > Art.....any opinions? Um, uh, I am having a REALLY bad week here...worst since the beginning of April. Let me look again..... We want %used so it should be used/total so Zandy is correct and Allen was not, nor me for that matter. Swap the formula to: (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 Art S. Kagel > Allen > > Zandy wrote: > > > > I'm not sure about this but isn't the formula for RA utilization > > equal to: > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > and not > > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > > as written in the script? > > > > Zandy > > > > ----------------------------------------------------------- > > > > Got questions? Get answers over the phone at Keen.com. > > Up to 100 minutes free! > > http://www.keen.com

↪ replying to Art S. Kagel

bwhite@deroyal.com — Mon, 21 Aug 2000 11:28:59 GMT — source: Informix-list mailing list archive (1991-1998)

What does it mean to have RA utilization above 100%? I've just increased my RA drastically (from 10 & 5 to 64 32). Bryan In article <39805D66.D659C1E2@bloomberg.net>, kagel@bloomberg.net wrote: > allenj wrote: > > > > I was wondering about this myself, because I was consistently getting > > a little more than 100% RA utilization. (100.63%) > > > > However, using your version of the formula, I get 99.36%. > > > > So I asked Art Kagel to confirm the formula that I used, and he said it > > was OK. > > > > We're probably splitting hairs here, but I wonder which one is correct? > > > > Art.....any opinions? > > Um, uh, I am having a REALLY bad week here...worst since the beginning > of April. Let me look again..... > > We want %used so it should be used/total so Zandy is correct and Allen > was not, nor me for that matter. Swap the formula to: > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > Art S. Kagel > > > Allen > > > > Zandy wrote: > > > > > > I'm not sure about this but isn't the formula for RA utilization > > > equal to: > > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > > > and not > > > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > > > as written in the script? > > > > > > Zandy > > > > > > ----------------------------------------------------------- > > > > > > Got questions? Get answers over the phone at Keen.com. > > > Up to 100 minutes free! > > > http://www.keen.com > Sent via Deja.com http://www.deja.com/ Before you buy.

↪ replying to bwhite@deroyal.com

allenj — Mon, 21 Aug 2000 13:58:49 GMT — source: Informix-list mailing list archive (1991-1998)

bwhite@deroyal.com wrote: > > What does it mean to have RA utilization above 100%? I've just > increased my RA drastically (from 10 & 5 to 64 32). So you are using the following formula for RA utilization: (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 and NOT: ((ixdaRA + idxRA + daRA) / RApgsused) * 100 And you're getting > 100%???? Jeez....and I thought my RA utilization was good at 99.36%........ <scratching head>.............. Allen > > Bryan > > In article <39805D66.D659C1E2@bloomberg.net>, > kagel@bloomberg.net wrote: > > allenj wrote: > > > > > > I was wondering about this myself, because I was consistently > getting > > > a little more than 100% RA utilization. (100.63%) > > > > > > However, using your version of the formula, I get 99.36%. > > > > > > So I asked Art Kagel to confirm the formula that I used, and he > said it > > > was OK. > > > > > > We're probably splitting hairs here, but I wonder which one is > correct? > > > > > > Art.....any opinions? > > > > Um, uh, I am having a REALLY bad week here...worst since the > beginning > > of April. Let me look again..... > > > > We want %used so it should be used/total so Zandy is correct and > Allen > > was not, nor me for that matter. Swap the formula to: > > > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > Art S. Kagel > > > > > Allen > > > > > > Zandy wrote: > > > > > > > > I'm not sure about this but isn't the formula for RA utilization > > > > equal to: > > > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > > > > > and not > > > > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > > > > as written in the script? > > > > > > > > Zandy > > > > > > > > ----------------------------------------------------------- > > > > > > > > Got questions? Get answers over the phone at Keen.com. > > > > Up to 100 minutes free! > > > > http://www.keen.com > > > > Sent via Deja.com http://www.deja.com/ > Before you buy.

↪ replying to bwhite@deroyal.com

Art S. Kagel — Tue, 22 Aug 2000 17:07:12 -0400 — source: Informix-list mailing list archive (1991-1998)

It means you are doing the calculation backwards. Art S. Kagel bwhite@deroyal.com wrote: > > What does it mean to have RA utilization above 100%? I've just > increased my RA drastically (from 10 & 5 to 64 32). > > Bryan > > In article <39805D66.D659C1E2@bloomberg.net>, > kagel@bloomberg.net wrote: > > allenj wrote: > > > > > > I was wondering about this myself, because I was consistently > getting > > > a little more than 100% RA utilization. (100.63%) > > > > > > However, using your version of the formula, I get 99.36%. > > > > > > So I asked Art Kagel to confirm the formula that I used, and he > said it > > > was OK. > > > > > > We're probably splitting hairs here, but I wonder which one is > correct? > > > > > > Art.....any opinions? > > > > Um, uh, I am having a REALLY bad week here...worst since the > beginning > > of April. Let me look again..... > > > > We want %used so it should be used/total so Zandy is correct and > Allen > > was not, nor me for that matter. Swap the formula to: > > > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > Art S. Kagel > > > > > Allen > > > > > > Zandy wrote: > > > > > > > > I'm not sure about this but isn't the formula for RA utilization > > > > equal to: > > > > (RApgsused / (ixdaRA + idxRA + daRA) ) * 100 > > > > > > > > and not > > > > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > > > > as written in the script? > > > > > > > > Zandy > > > > > > > > ----------------------------------------------------------- > > > > > > > > Got questions? Get answers over the phone at Keen.com. > > > > Up to 100 minutes free! > > > > http://www.keen.com > > > > Sent via Deja.com http://www.deja.com/ > Before you buy.

Related threads

Glossary

onstat