Re: Calculating SIIS% as MSUs or MIPS

[Scott Chapman]

I agree. I don't think you can readily deduce the potential absolute CPU 
capacity impact from the SIIS % number. That's no doubt why IBM gave general 
guidance on how serious you should consider the problem rather than saying "oh 
this is costing you this many CPU seconds/MSUs//MIPS". My recommendation is to 
also consider the actual CPU consumption during the periods where there is a 
high SIIS% issue. I've seen cases where the SIIS % is quite high but the actual 
CPU consumption of the LPAR is relatively minor (e.g. perhaps a sysprog test 
LPAR or a relatively idle production LPAR) meaning that the net savings will be 
also relatively minor in the overall scheme of things. 

Scott Chapman

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Adam Johanson]

A previous poster wrote:

> siis% is number of instructions that suffered SIIS in a given period

But I wonder about this. The percentage in "SIIS percentage" technically isn't 
a percentage of instructions, but rather percentage regarding the source of L1 
i-cache directory writes. Looking at how SIIS percentage is calculated* on the 
z15 shows that it's:

E164 / B2 * 100%

where:
E164: Extended counter 164
B2: Basic counter 2

Extended counter 164:

Counter 164 – A directory write to the Level-1
Instruction cache directory where the returned cache
line was sourced from an On-Chip Level-3 cache
with intervention.

Basic Counter 2: 

Level-1 I-cache directory-write count


... so the percentage deals with how often the L1 i-cache had to be refreshed 
from L3, indicating a possible SIIS violation with the thought that it's 
because the i-cache had to be refreshed because it was changed.

Correlating this down to the "number of instructions that suffered SIIS" looks 
like it could be plausible from a high-level perspective, but I'm not sure. If 
an instruction stalls out because of its L1 i-cache line having to be 
refreshed, then it seems like this would correlate, but with all that happens 
in the pipeline, it just doesn't feel like this would always correlate 1:1.

I'm thinking of things like when cache lines are pre-fetched. If during 
pre-fetch, a cache line had to be sourced from L3 because it changed since its 
last use, then (since that happens ahead of time) one particular instruction 
couldn't be singled out as suffering from the refresh from L3.

Like I mentioned at the start of the post, I'm not saying this is straight up 
wrong, I just wonder about it.


Source: 
https://www.ibm.com/support/pages/system/files/inline-files/Identifying%20SIIS%20Inefficiency%20by%20Using%20CPU%20MF%20Counters%20slides%20Updated%20May%202022_0.pdf

===
Adam Johanson
Broadcom Mainframe Software Division

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Andreas von Imhof]

Dump 70,72,113 from every system on the z server

Run the following build: MYIN is FB 80

%LET MACKEEP=  MACRO _XLA113  _XLA113F  % ;
%UTILBLDP(BUILDPDB=NO,USERADD=7072 113,
  OUTFILE=MYIN,
  WANTSMF=70.1 113.1,  
  INCLAFTR=ASUM70PR ASUM113);  
  %INCLUDE MYIN; RUN;  

set dagje.ASUM113 (obs=100);

For a z15 the following calc when listing
z15 E164 / B2 * 100% I Writes sourced with L3 intervention / I Writes

siis = (extnd164 / basic02) / 100 ;

SMF70CIN gives you CP or IIP
CECSER for the CEC if you have more than one
SMF70CPA the SU-SEC (see I like MSU)
MIPSEXEC for the executed MIPS (only if you are old like me)

Barry makes it all very easy for us.

PS if SAS / MXG is not running on your clients system but on your own system 
then please check for SW license issues 1st.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[shivang sharma]

I will be using SAS MXG . I am fine with representing it in MSUs . But how
should i derive it. Will it be siis%
of the lpar total MSUs used in that interval?

On Fri, 5 May 2023 at 2:50 PM, Andreas von Imhof <
02ba5d449f94-dmarc-requ...@listserv.ua.edu> wrote:

> Using type 30's is not my 1st choice. Need to calculate capture ratio's,
> preferably by workload type. Cannot remember if DB2 type enclaves are
> captured here or it is all lumped into DB AS.
>
> I think that Shivang wants a total view. The document from John Burg
> spells it out quite clearly.
> eg page 4: z16 E170 / B2 * 100% I Writes sourced with L2 intervention / I
> Write
>
> I would take type 70's, match the intervals with the 113's and calculate
> the percentage.
> Never use MIPS, use HW or SW MSU (clearly state which one you are using).
> Why MSU? All contracts / pricing are based on this.
> MSU is also very easy to match to type 72's. Type 72's makes reporting
> easy peasy.
>
> Shivang, what tooling are you going to use to do this?
>
> Andreas von Imhof
> Capacity & Performance z/OS - zCAP
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Andreas von Imhof]

Using type 30's is not my 1st choice. Need to calculate capture ratio's, 
preferably by workload type. Cannot remember if DB2 type enclaves are captured 
here or it is all lumped into DB AS.

I think that Shivang wants a total view. The document from John Burg spells it 
out quite clearly.
eg page 4: z16 E170 / B2 * 100% I Writes sourced with L2 intervention / I Write

I would take type 70's, match the intervals with the 113's and calculate the 
percentage.
Never use MIPS, use HW or SW MSU (clearly state which one you are using). 
Why MSU? All contracts / pricing are based on this.
MSU is also very easy to match to type 72's. Type 72's makes reporting easy 
peasy.

Shivang, what tooling are you going to use to do this?

Andreas von Imhof
Capacity & Performance z/OS - zCAP 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Martin Trübner]

Where is the problem ?


siis% is number of instructions that suffered SIIS in a given period

MIPS is number of instructions per second

you have cpu-seconds used in a given period or the number of 
instructions executed in that given period



 you don't want us do the math for you, or?


Martin


Am 05.05.23 um 11:05 schrieb shivang sharma:

I am not looking to know how to calculate siis% from smf113 . I am looking
for how to convert SIIS% in terms of MSUs or MIPS.

On Fri, 5 May 2023 at 2:33 PM, Martin Trübner <
047eec287bd9-dmarc-requ...@listserv.ua.edu> wrote:


the May 2022 update has the numbers you are looking for

Am 04.05.23 um 23:46 schrieb shivang sharma:

Hi ,


Every IBM document discribes this as % . I am trying to get the SIIS%

value

in terms of MIPS or MSUs

On Fri, 5 May 2023 at 1:49 AM, Sri h Kolusu   wrote:


Shivang,

Check these links.




https://www.ibm.com/support/pages/identifying-%E2%80%-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters

<

https://www.ibm.com/support/pages/identifying-%E2%80%9Cstore-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters





https://www.ibm.com/support/pages/system/files/inline-files/CPUMF_Implementation_Details.pdf

Thanks,
Kolusu

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send emailtolists...@listserv.ua.edu   with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send emailtolists...@listserv.ua.edu   with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[shivang sharma]

I am not looking to know how to calculate siis% from smf113 . I am looking
for how to convert SIIS% in terms of MSUs or MIPS.

On Fri, 5 May 2023 at 2:33 PM, Martin Trübner <
047eec287bd9-dmarc-requ...@listserv.ua.edu> wrote:

> the May 2022 update has the numbers you are looking for
>
> Am 04.05.23 um 23:46 schrieb shivang sharma:
> > Hi ,
> >
> >
> > Every IBM document discribes this as % . I am trying to get the SIIS%
> value
> > in terms of MIPS or MSUs
> >
> > On Fri, 5 May 2023 at 1:49 AM, Sri h Kolusu  wrote:
> >
> >> Shivang,
> >>
> >> Check these links.
> >>
> >>
> >>
> https://www.ibm.com/support/pages/identifying-%E2%80%-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters
> >> <
> https://www.ibm.com/support/pages/identifying-%E2%80%9Cstore-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters
> >
> >
> >>
> >>
> https://www.ibm.com/support/pages/system/files/inline-files/CPUMF_Implementation_Details.pdf
> >>
> >> Thanks,
> >> Kolusu
> >>
> >> --
> >> For IBM-MAIN subscribe / signoff / archive access instructions,
> >> send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN
> >>
> > --
> > For IBM-MAIN subscribe / signoff / archive access instructions,
> > send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Martin Trübner]

the May 2022 update has the numbers you are looking for

Am 04.05.23 um 23:46 schrieb shivang sharma:

Hi ,


Every IBM document discribes this as % . I am trying to get the SIIS% value
in terms of MIPS or MSUs

On Fri, 5 May 2023 at 1:49 AM, Sri h Kolusu  wrote:


Shivang,

Check these links.


https://www.ibm.com/support/pages/identifying-%E2%80%-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters





https://www.ibm.com/support/pages/system/files/inline-files/CPUMF_Implementation_Details.pdf

Thanks,
Kolusu

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email tolists...@listserv.ua.edu  with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Sri h Kolusu]

>>Every IBM document discribes this as % . I am trying to get the SIIS% value 
>>in terms of MIPS or MSUs

Shivang,

Did you look at SMF30COUNT ? If it is set, you can get address space MSU 
consumptions.

Alternatively, you can covert the CPU seconds(SMF30) to MSUs and MIPS

https://www.longpelaexpertise.com.au/ezine/CPUtoMSUMIPS.php



Thanks,
 Kolusu


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[shivang sharma]

Hi ,


Every IBM document discribes this as % . I am trying to get the SIIS% value
in terms of MIPS or MSUs

On Fri, 5 May 2023 at 1:49 AM, Sri h Kolusu  wrote:

> Shivang,
>
> Check these links.
>
>
> https://www.ibm.com/support/pages/identifying-%E2%80%-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters
> 


>
>
> https://www.ibm.com/support/pages/system/files/inline-files/CPUMF_Implementation_Details.pdf
>
> Thanks,
> Kolusu
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Calculating SIIS% as MSUs or MIPS

[Sri h Kolusu]

Shivang,

Check these links.

https://www.ibm.com/support/pages/identifying-%E2%80%9Cstore-instruction-stream%E2%80%9D-siis-inefficiency-using-cpu-mf-counters

https://www.ibm.com/support/pages/system/files/inline-files/CPUMF_Implementation_Details.pdf

Thanks,
Kolusu

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Calculating SIIS% as MSUs or MIPS

[shivang sharma]

Hi All ,

SMF113 provides metrics to calculate siis% . At one customer shop we have
been asked to provide estimated SIIS MSUs or MIPS . Can somebody point me
an easy way to derive this.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN