* copyright (C) 1984-2019 merrill consultants dallas texas usa

your SMF data so I can validate and update with data.

Thanks to Rudolf Sauer, T-SYSTEMS, GERMANY.

Change 31.161 Cosmetic. Three UNINITIALIZED variable messages removed

VMACRMFV and debugging PUTLOG COL= N statement removed.

Aug 1, 2013
Change 31.160 Support for Velocity Software new segments that create

EXXAMCSH these new datasets (This MAY have been zVPS 4.2.)

EXXAMCTY

EXXAMHPP DDDDDD DATASET DESCRIPTION INFILE

EXXAMSPT

EXXMUDIO XAMCSH XMCPSHAR CPSHAR XAMSYS

EXXMVCPU XAMCTY XMCPTYPE CPTYPE XAMSYS

IMACXAM XAMHPP XMIODHPP IODHPP XAMDEV

VMACXAM XMUDIO XMUCDDIO UCDDIO XAMTCP

VMXGINIT XMVCPU XMVSICPU VSICPU XAMTCP

Aug 20, 2013 XAMSPT XMSYTSPT SYTSPT XAMSYS

Aug 21, 2013 and add new variables to many existing XAM datasets.

-All datasets are created and all tokens defined.

-Large VM systems can create multiple XAMSYS records for

each interval, but since only that first record contains

all of the 'single-instance-per-interval' segments that

are all output into dataset XAMSYS, those additional

records created an observation in XAMSYS with all values

a missing value. Now, only that first record is output

in XAMSYS (and, of course, those other multiple-instance

segments are output to their unique MXG dataset.

-VSICPU: There is one set of counters per Virtual CPU, and

SEGLEN is used to calculate the number of segments.

-SYTCPM: Rather than creating one obs for each channel in

the XMSYTCPM dataset, the 256 channel busy variables

PCTCHPBUSY00-PCTCHPBUSYFF are now output in XAMSYS and

dataset XMSYTCPM is no longer created, and references to

to dddddd token SYTCPM are removed from VMACXAM and the

EXSYTCPM member is deleted.

-The SYTCPM segment NCHAN field does not count the number

of channel busy metrics in the record; instead, it is the

the highest channel number. And while SEGLEN does count

the actual number of metrics in the record, it is wrong

in current XAM records with an extra four bytes:

a. NCHANS='FF'x (255), SEGLEN=1140 => 257 actual metrics

b. NCHANS='FB'x (251), SEGLEN=1124 => 253 actual metrics

and that 257 value raised an ARRAY SUBSCRIPT error with

only 256 possible metrics expected. While Barton will

fix SYTCPM so the SEGLEN is correct, this

change/circumvention simply increases the array size to

257 but only the first 256 variables are kept, so you

won't have to change MXG when Barton makes his change.

Thanks to Jim Horne, Lowe's, USA.

Thanks to Douglas C. Walter, Citigroup, USA.

Change 31.159 Mostly cosmetic; HOLDxxxx variables are DROPped and the

ANAL42DS TASKTYPE typo was corrected to TYPETASK.

Jul 30, 2013
Change 31.158 Circumvention for added fields in TMQQAA that caused the

VMACTMMQ 2nd and subsequent observations from the same record to

Jul 30, 2013 be misaligned.
Change 31.157 Examples in JCLUOW, JCLMQMCI & ASUMUOW were inconsistent

JCLUOW and in some cases incorrect, and created unneeded data.

JCLMQMCI JCLUOW - Creates CICSTRAN,DB2ACCT, then PDB.ASUMUOW

ASUMUOW JCLMQMCI - Adds MQMACCT,MQMACCTQ, creates PDB.ASUMUOW.

Jul 29, 2013

Thanks to Thomas Kelman, Xerox, USA.

====== Changes thru 31.156 were in MXG 31.05 dated Jul 29, 2013=========
Change 31.156 -MACROs _N70, _N72, _S70, and S72 are created so that you

VMAC7072 can create only datasets from ID=70 or ID=72 RMF records.

Jul 25, 2013 (Members TYPE7072/TYPS7072 process both records because

early RMF required passing CPUTYPE from the 70 to the 72

so that CPU could be calculated from service units with a

table lookup to get SU_SEC. Yep, I said EARLY!)

While most will always create both suites of datasets,

you can use these new macro tokens to process only one of

the two record; for example:

//SYSIN DD *

%LET MACKEEP=

_N72

MACRO _S72 %

;

%INCLUDE SOURCLIB(TYPS7072);

will only create the PDB.TYPE70xx datasets:

_N72 Nulls the creation and OUTPUT of TYPE72xx datasets

_S72 Blanks/Nulls the SORTs of TYPE72xx datasets.

-MACRO _DROP70 lists the TYPE70 per-engine variables

so you can ignore them using

%INCLUDE SOURCLIB(VMAC7072);

PROC PRINT DATA=PDB.TYPE70 (DROP- _DROP70);

or drop them from a new dataset using:

%INCLUDE SOURCLIB(VMAC7072);

DATA WANT;SET PDB.TYPE70; DROP _DROP70;

-TYPE70 only keeps the first 102 per-engine variables.

Dataset TYPE70EN has all 255 possible per-engines.

Change 31.155 -Protection for INVALID DATA FOR RVPERCEN when IBM stored

VMACEDGR '***' EBCDIC for the numeric VOLUME*PERCENTAGE*FULL.

Jul 25, 2013 The double question-mark-modifier was added to RVPERCEN

INPUT to suppress the ERROR message and the hex dump.

With or without this protection, the value in RVPERCEN

is a missing value. If I can find out what those three

asterisks mean, I'll update this note.

-Change 31.118 failed to input RVMEDTY causing subsequent

RVxxxxx variables to be trashed in EDGRXEXT dataset.

Thanks to Paul Volpi, UHC, USA.

Change 31.154 By default, ASUMHSM summarizes all HSM systems, but you

ANALHSM can use MACRO _HSMPLEX HSMPLEX=SYSTEM; % in a

ASUMHSM %LET MACKEEP= in your SYSIN to create a separate summary

TRNDHSM for each system, ANALHSM was updated to allow HSMPLEX to

Jul 25, 2013 be used to report each SYSTEM/LPAR, and variable HSMPLEX

(default is blank) is added to the SUMBY= in TRNDHSM.

Thanks to Rick Ralston, Humana, USA.
Change 31.153 Support for z/OS 2.1: COMPATIBLE, VALIDATED WITH IBM DATA

BUIL3005 -z/OS 2.1 Compressed Log Stream can only be read with

BUILD005 IFASMFDL running on a z/OS 2.1 system with zEnterprise

EXT11941 Data Compress (zEDC) Express. Otherwise, IFASMFDL will

EXT11942 fail with Return Code 4, unless you have enabled

EXT11943 SOFTINFLATE option of IFASMFDL, which will decompress,

EXT11944 but at higher CPU cost than zEDC Express.

EXT1194L -"You can specify LRECL=32760 instead of 32767" (default)

EXT11971 is finally now stated in the SMF manual, but 32767 should

EXTY42VC NEVER have been used for the LRECL. It can never be

EXTY9036 greater than 32760 and 32767 has caused many programs to

FORMATS fail (BUT NOT SAS!!). And, of course, IBM is reluctant

VMAC104 to change ANY default, even those they recommend that you

VMAC113 should change!

VMAC119 -New SMFPRMxx PERMFIX parameter sets minimum (1MB) to

VMAC1415 maximum (2GB) page fixed real storage that is set aside

VMAC23 for zEDC Express decompression services for compressed

VMAC26J2 SMF data. NOPERMFIX default, but up to 2MB pages can be

VMAC30 fixed for zEDC Express.

VMAC30 -New SMF30 Instruction Counters variables are added to ALL

VMAC42 MXG datasets derived from SMF 30s that contain CPUTCBTM:

VMAC6 TYPE30_V,SMFINTRV,TYPE30_4,TYPE30_5,PDB.STEPS,PDB.JOBS;

VMAC60 -IBM now recommends these 30 hardware instruction counters

VMAC7072 should be used for chargeback instead of processor time:

VMAC71 SMF30_INST_CP_TASK =SMF30*INST*CP*TASK

VMAC73 SMF30_INST_CP_NONPREEMPTSRB =SMF30*INST*CP*NONPREEMPTSRB

VMAC74 SMF30_INST_CP_PREEMPTSRB =SMF30*INST*CP*PREEMPTSRB

VMAC90 SMF30_INST_OFFLOAD =SMF30*INST*OFFLOAD

VMACDCOL SMF30_INST_OFFLOADONCP =SMF30*INST*OFFLOADONCP

VMXGINIT SMF30_INST_CP_ENCLAVE =SMF30*INST*CP*ENCLAVE

EXTY749 SMF30_INST_OFFLOAD_ENCLAVE =SMF30*INST*OFFLOAD*ENCLAVE

SMF30_INST_OFFLOADONCP_ENCLAVE=SMF30*INST*OFFLOADONCP*ENCLAVE

SMF30_INST_CP_DEPENC =SMF30*INST*CP*DEPENC

SMF30_INST_OFFLOAD_DEPENC =SMF30*INST*OFFLOAD*DEPENC

SMF30_INST_OFFLOADONCP_DEPENC =SMF30*INST*OFFLOADONCP*DEPENC

Jul 23, 2013 -Sums of subsets of these counts create new variables that

correspond to the existing CPU time variables:

CPU_INST = Total ALL ENGINES INSTRUCTION COUNT

CPUTCBTM_INST= SMF30_INST_CP_TASK+ /*CPT_INST*/

SMF30_INST_CP_PREEMPTSRB+

SMF30_INST_OFFLOADONCP+

SMF30_INST_CP_ENCLAVE+

SMF30_INST_OFFLOADONCP_ENCLAVE+

SMF30_INST_CP_DEPENC+

SMF30_INST_OFFLOADONCP_DEPENC;

CPUSRBTM_INST= SMF30_INST_CP_NONPREEMPTSRB; /*CPS_INST*/

CPUASRTM_INST= SMF30_INST_CP_PREEMPTSRB; /*ASR_INST*/

CPUENCTM_INST= SMF30_INST_CP_ENCLAVE+ /*ENC_INST*/

SMF30_INST_OFFLOADONCP_ENCLAVE;

CPUDETTM_INST= SMF30_INST_CP_DEPENC+ /*DET_INST*/

SMF30_INST_OFFLOADONCP_DEPENC;

-IBM notes originally had ONLY the _OFFLOAD counter for

the zIIP+zAAP counts, but data with _OFFLOAD zero and

non-zero _OFFLOAD_DEPENC confirmed the note was wrong,

and IBM has confirmed these calculations are correct:

CPUZIPTM_CPUIFATM_INST=SMF30_INST_OFFLOAD-

SMF30_INST_OFFLOAD_DEPENC-

SMF30_INST_OFFLOAD_ENCLAVE;

(i.e. total instructions on zIIP/zAAP)

CPUTM_INST =CPU_INST -

CPUZIPTM_CPUIFATM_INST;

(i.e. total instructions on CP engines).

-TYPE30_V,_4,_5,_6 new variable:

SMF30_RCMTADJN='NOMINAL*CPU*RATE*ADJUSTMENT'

These new INSTRUCTION COUNTs require SMF30COUNT be

specified in SMFPRMxx, and HIS (SMF 113) must have

enabled ehe Basic Counter Set.

-SMF 6 Printway record adds new JOB ACCOUNTing fields in

extended mode (INDC=7) records. See Change 32.236.

-TYPE1415 new SM14DSTYPE identifies Extended Format

Type 0/1, or Version 2 decoded by $MG014EF format.

-TYPE23, new variables:

SMF23CWN='COMPRESSED*LOG*BLOCKS*WRITTEN'

SMF23NCN='NON-COMPRESSED*LOG BLOCKS*WRITTEN'

SMF23PFH='PERMFIX*HWM USED*BY SMF'

SMF23PFM='PERMFIX*MAXIMUM*ALLOWED'

SMF23PFT='PERMFIX*STORAGE*USED*BY*SMF'

-TYPE26J2, New Sub-subtype, not documented in SMF manual

yet, will update this note when the new manual is out.

-TYPE42 VCC Virtual Concurrent Copy segment (previously

existed, but overlooked) creates new TYPE42VC dataset:

S42VCCTK='TRACKS*USING*CONCURRENT*COPY'

S42VCDSP='TRACKS*USING*VIRTUAL*CCOPY'

S42VCEIT='INITIALIZATION*END*TODSTAMP'

S42VCID ='VCC*LOGICAL*SESSION*ID'

S42VCRQS='REQUEST*TYPE VCC*=VIRT CONCUR COPY'

S42VCSET='SESSION*END*TODSTAMP'

S42VCSSL='LENGTH*OF*SSIDS*FOR SESSION'

S42VCSSN='NUMBER*OF*SSIDS*FOR SESSION'

S42VCSSO='OFFSET*TO FIRST*SSID'

S42VCSST='SESSION*START*TODSTAMP'

S42VCTS ='TERMINATION*STATUS*N=NORMAL*A=ABNORM'

-TYPE60 new SMF60ELP elapsed duration which can be

subtracted from SMFTIME to get the START time.

-TYPE70, new variables:

SMF70MCP =MAXIMUM*CPU*ADDRESS*AVAILABLE

SMF70ICP =HIGHEST*CPUID*INSTALLED*AT IPL

SMF70CCP =HIGHEST*CPUID*CURRENTLY*INSTALLED

SMF70CPA_ACTUAL =PHYSICAL*CPU*ADJUSTMENT*FACTOR

SMF70CPA_SCALING_FACTOR =SMF70CPA_ACTUAL*SCALING*FACTOR

SMF70UIW='SHOULD*USE*INITIAL*WEIGHT?'

SMF70HW_CAP_LIMIT=ABS LIMIT*PARTITION*USAGE*IN NR ENGINES

Note: field contains hundredths of "CPUs", but MXG

converts to count "CPUs", so 1.5 would be one

and one half engines of type SMF70CIX.

Note: APAR OA40539 added this field to z/OS 1.13.

-TYPE71, new variables:

SMF71S7M='MIN SHARED*PAGE GROUPS*ON SCM'

SMF71S7X='MAX SHARED*PAGE GROUPS*ON SCM'

SMF71S7A='AVG SHARED*PAGE GROUPS*ON SCM'

-TYPE72, new dataset TYPE725N='GRS QSCAN STATISTICS':

R725QSAS='ADDRESS SPACE ID'

R725QSJN='NAME OF THE JOB'

R725QSRC='START*RESUME*REQUESTS'

R725QSRQ='SUM OF SQUARES*RESOURCES*RETURNED'

R725QSRR='RESOURCES*RETURNED*FOR THESE*REQUESTS'

R725QSSC='GQSCAN*ISGQUERY*SPECIFIC*REQUESTS'

R725QSSN='SERVICE CLASS NAME'

R725QSSP='SERVICE CLASS PERIOD'

R725QSST='ADDRESS SPACE STOKEN'

R725QSTI='EXECUTION*TIME IN*GRS*THESE*REQUESTS'

R725QSTQ='SUM OF SQUARES*EXECUTION*TIME'

-TYPE73, new variables:

SMF73MSC='CHANNEL*PATH*POWER'

SMF73SPD='CHANNEL*SPEED*BITS PER*SECOND'

-TYPE74, new variables:

FICONDEV='FICON*DEVICE?'

AVG74CUQ='AVERAGE*CU*QUEUE*PER IO'

SMF74CUQ='CONTROL*UNIT*QUEING*TIME'

SMF74NM2='DEVICE*NUMBER'

SMF74SCS='SUBCHANNEL*SET*ID'

-TYPE74CF, new variables:

R744GTSC='CF*SCM*MAY BE USED*EXTENSIONS'

R744GFSC='FREE*CF*SCM*MEMORY'

R744GISC='SCM*INCREMENT'

-TYPE74MO, new dataset: STORAGE CLASS MEMORY

See Change 33.155. SCM data is now in TYPE74ST dataset.

R744MAEC='SCM*AUXILIARY*ENABLED*COMMANDS'

R744MALG='SCM*ALGORITHM*TYPE'

R744MCPI='CHANNEL*PATH*IDENTIFIER'

R744MEMA='EST MAX*ASSIGNED*AUGMENTM SPACE'

R744MEME='EST MAX*LIST*ELEMENTS*IN SCM'

R744MEML='EST MAX*LIST*ENTRIES*IN SCM'

R744MENE='EXISTING*LIST*ELEMENTS*IN SCM'

R744MENL='EXISTING*LIST*ENTRIES*IN SCM'

R744MFAU='FIXED*AUGMENTED*SPACE'

R744MIUA='AUGMENTED*SPACE*IN USE*BY THIS STRUCTURE'

R744MIUS='SCM*IN USE*BY THIS*STRUCTURE'

R744MMBE='MAX*LIST ELEMENTS*PER SCM*BUFFER'

R744MMBL='MAX*LIST ENTRIES*PER SCM*BUFFER'

R744MRBT='SCM*READ*BYTES*TRANSFERRED'

R744MRFC='SCM*READ OPS*LIST*REFERENCE'

R744MRPC='SCM*READ OPS*PREFETCH OP'

R744MRSQ='SQUARES OF*R744MRST'

R744MRST='SCM*READ OPS*SERVICE*TIME'

R744MSLR='PCT*LIST COUNTS*LOWER*REGULATOR'

R744MSLT='PCT*LIST COUNTS*LOWER*THRESHOLD'

R744MSMA='MAX*SCM*STRUCTURE*CAN USE'

R744MSRL='SCM*REFERENCES*LIST*STRUCTURE'

R744MSRM='SCM*REFERENCES*MIGRATION'

R744MSRR='SCM*REFERENCES*LIST*HASHING'

R744MSUR='PCT*LIST COUNTS*UPPER*REGULATOR'

R744MSUT='PCT*LIST COUNTS*UPPER*THRESHOLD'

R744MSWC='SCM LIST WRITES'

R744MWBT='SCM*WRITE*BYTES*TRANSFERRED'

R744MWSQ='SQUARES OF*R744MWST'

R744MWST='SCM*WRITE OPS*SERVICE*TIME'

-TYPE749, new dataset: RMF III PCIE STATISTICS, but zero

observations are created, awaiting test data; it's very

possible multiple new dataset may ultimately be created.

All variables are created for QA comparisons.

-TYPE90. New subtype 36 creates new dataset

dddddd dataset Description ST

TY9036 TYPE9036 SET CON COMMAND 36

Subtype 35 was not documented.

-TYPE99EH. New variables:

S99EEHMAXAFF='MAXIMUM*AFFINITY*INDEX'

S99EEHMAXCPU='MAXIMUM*CPU*ID'

-New SMF104 record captures CIM data for AIX, LINUX, and

WINDOWS, creating 37 new TYP104nn datasets with nn the

record subtype. Some of the memory metrics in the raw

record are in bytes, or KiloBytes, or MegaBytes: ALL are

converted in MXG variables to bytes and then formatted

MGBYTES to document they contain memory metrics and to

display in B/KB/MB/GB/TB etc. Some "accumulated" fields

are documented, but until data is available, I can't tell

if they really need deaccumulation or not, which would be

an update to their _ST104xx dataset sort macro.

dddddd dataset Description ST

T10401 TYP10401 AIX_ACTIVE MEMORY EXPANONMETRICS 01

T10402 TYP10402 AIX_PROCESSOR METRICS 02

T10403 TYP10403 AIX_COMPUTER SYSTEM METRICS 03

T10404 TYP10404 AIX_DISK METRICS 04

T10405 TYP10405 AIX_NETWORK PORT METRICS 05

T10406 TYP10406 AIX_FILE SYSTEM METRICS 06

T10407 TYP10407 AIX_MEMORY METRICS 07

T10408 TYP10408 AIX_OPERATING SYSTEM METRICS 08

T10409 TYP10409 AIX PROCESS METRICS 09

T10410 TYP10410 AIX_SHARED ETHERNET ADAERMETRICS 10

T10411 TYP10411 AIX_ACTIVE MEMORY SHARIMETRICS 11

T10412 TYP10412 AIX_VIRTUAL TARGET DEVICEMETRICS 12

T10420 TYP10420 LINUX_IP PROTOCO LENDPOT METRICS 20

T10421 TYP10421 LINUX_LOCAL FILE SYSTE METRICS 21

T10422 TYP10422 LINUX_NETWORK PORT METRICS 22

T10423 TYP10423 LINUX_OPERATING SYSTE METRICS 23

T10424 TYP10424 LINUX_PROCESSOR METRICS 24

T10425 TYP10425 LINUX_UNIX PROCESS METRICS 25

T10426 TYP10426 LINUX_STORAGE METRICS 26

T10430 TYP10430 LINUX_KVM METRICS 30

T10431 TYP10431 LINUX_XEN METRICS 31

T10440 TYP10440 LINUX_IP PROTOCOL ENDPOT METRICS 40

T10441 TYP10441 LINUX_LOCAL FILE SYSTE METRICS 41

T10442 TYP10442 LINUX_NETWORK PORT METRICS 42

T10443 TYP10443 LINUX_OPERATING SYSTEM METRICS 43

T10444 TYP10444 LINUX_PROCESSOR METRICS 44

T10445 TYP10445 LINUX_UNIX PROCESS METRICS 45

T10446 TYP10446 LINUX_STORAGE METRICS 46

T10450 TYP10450 LINUX_ZCEC METRICS 50

T10451 TYP10451 LINUX_ZLPAR METRICS 51

T10452 TYP10452 LINUX_ZCHANNEL METRICS 52

T10453 TYP10453 LINUX_ZECKD METRICS 53

T10460 TYP10460 WINDOWS_LOCAL FILESYSTEM METRICS 60

T10461 TYP10461 WINDOWS_NETWORK PORT METRICS 61

T10462 TYP10462 WINDOWS_OPERATING SYSTEM METRICS 62

T10463 TYP10463 WINDOWS_PROCESSOR METRICS 63

T10464 TYP10464 WINDOWS_STORAGE METRICS 64

-TYPE113 Subtype 1 is documented and variables defined,

but will not be read until data records exist for the

validation.

-TYPE119. Six new datasets, with all variables defined,

but no observations will be created; INPUT is bypassed

until test records are available for validation.

dddddd Dataset Description

T11941 TYP11941 SMC-R LINK GROUP STATISTICS

T1194L TYP1194L SMC-R LINK SPECIFIC STATISTICS

T11942 TYP11942 SMC-R LINK STATE START

T11943 TYP11943 SMC-R LINK STATE END

T11944 TYP11944 RDMA RNIC INTERFACE STATISTICS

T11971 TYP11971 FTP DAEMON CONFIGURATION

-TYPEDCOL.

New variable in DCOLDSET:

DCDXPSEV='PS*EXTND*FORMAT*VERSION'

New variable in DCOLDC:

DDCRMODE='VSAM*SMB*RMODE31*VALUE'

New variable in DCOLDC:

DBSEPNM ='SEPARATION*NAME'

THIS CHANGE WAS IN MXG 31.04 BUT WAS NOT DOCUMENTED.
Change 31.152 TYPE70 corrections/enhancements to zIIP and zAAP metrics.

VMAC7072 Variables ZIPUPTM/IFAUPTM, Up-Time, now excludes parked

Jul 24, 2013 time (like CP engines in PDB.TYPE70 and like ALL engines

in ASUM70PR-created datasets - it was only the specialty

engines in PDB.TYPE70 that still included parked time.).

so variables PCTZIPBY and PCTIFABY will be larger.

New counts of the Online-but-not-Parked zIIP/zAAP engines

and clarification of the variables with INSTALLED counts

are added, as are online and parked durations for the

zIIPs and zAAPs:

Engine INSTALLED AVAILABLE UP-TIME ONLINE PARKED
IFA NRIFAS IFACPUS IFAUPTM IFAONTTM IFAPATTM

SMF70IFA

PARTNIFA
ZIP NRZIPCPU ZIPCPUS ZIPUPTM ZIPONTTM ZIPPATTM

SMF70SUP

PARTNZIP
CP NRCPCPU NRCPUS CPUUPTM SMF70ONT SMF70PAT

PARTNCPU
New and revised labels:

PARTNICF='INSTALLED*NUMBER OF*ICF*ENGINES'

PARTNIFA='INSTALLED*NUMBER OF*ZAAP*ENGINES'

PARTNIFL='INSTALLED*NUMBER OF*IFL*ENGINES'

PARTNZIP='INSTALLED*NUMBER OF*ZIIP*ENGINES'

NRCPUS ='ONLINE*NON-PARKED*CP ENGINES'

IFACPUS ='ONLINE*NON-PARKED*IFA ENGINES'

ZIPCPUS ='ONLINE*NON-PARKED*ZIP ENGINES'
CPUUPTM ='CP ENGINE*AVAILABLE*(UP) TIME'

IFAUPTM ='IFA ENGINE*AVAILABLE*(UP) TIME'

ZIPUPTM ='ZIP ENGINE*AVAILABLE*(UP) TIME'
SMF70PAT='CP*SMF70PAT*PARKED*TIME'

IFAPATTM='IFA*SMF70PAT*PARKED*TIME'

ZIPPATTM='ZIP*SMF70PAT*PARKED*TIME'
SMF70ONT='CP*SMF70ONT*ONLINE*TIME'

IFAONTTM='IFA*SMF70ONT*ONLINE*TIME'

ZIPONTTM='ZIP*SMF70ONT*ONLINE*TIME'
zIIP Example: PARTNZIP=12 ZIPCPUS=8.01 DURATM=15:00
|--------------------ZIPONTTM 2:59:59.74---------------------------|

| |

|-----------ZIPUPTM 2:00:09.21------------|---ZIPPATTM 59:50.53---|

| |

|-ZIPACTTM 1:16:48.85-|-ZIPWAITM 42:33.59-|

Change 31.151 -Support for zVM 6.3 MONWRITE. INCOMPAT DUE TO MXG CODE,

VMACVMXA which didn't SKIP new data compatibly added to SYTCUP and

Feb 20, 2013 and didn't correctly detect the extra 32 bytes in PRCAPM

Mar 26, 2013 subtype 8, and printed ERROR PROBABLE DATA LOSS DUE TO...

Apr 16, 2013 and BROKEN CONTROL RECORD: NO 8709X, with large _N_=.

VMXGINIT This change was made in VMACVMXA at 30.30.

Apr 24, 2013 -Before _ESYTCUP statement near line 11567, insert:

Jul 23, 2013 SKIP=CALCPULN-40;IF SKIP GT 0 THEN INPUT +SKIP @;

This change was made in VMACVMXA at 31.02.

-Replace (For 5.10 Record showstopper):

IF PRCAPMCT IN (4,8) THEN DO; line 19028 with

IF SUBSTR(PRCAPMV,1,1) NE 'F0'X THEN DO;

-Two instances of ENDTIME=.; lines 8633,8907 are now

comments, just in case, as they are archaic and not

needed (to force detection of a new interval) but in

some cases caused ENDTIME to be missing in startup

records.

-With those changes, records were read with no execution

errors with MXG 31.02; support was NOT announced.

-These new variables and/or datasets are now created:

SKIP LEN _N_ RECSTART

-Dataset VXSYTPRP (0.02) new variables: 12 152 101 373

CALENTMT='SCALED*PORTION*VERTICAL*ENTITLED'

PFXPRKWT='PARKED*WAIT*TIME'

-Dataset VXSYTASG (0.06) new variables: 8 92 106 493

CALENTMT='SCALED*PORTION*VERTICAL*ENTITLED'

PFXPRKWT='PARKED*WAIT*TIME'

-Dataset VXSYTUSR (0.08) new variables: 8 108 106 657

RLOIB ='INBOUND*LIVE*GUEST*RELOCATIONS'

RLOOB ='OUTBOUND*LIVE*GUEST*RELOCATIONS'

-Dataset VXSYTCUP (0.16) new variables: 8 1216 107 1605

LCXCCWT ='CURRENT*LPAR*WEIGHT'

LCXCTYCP='CPU*TYPE*CAP'

CALCAPV ='MAX*PHYSICAL*CPS*THIS TYPE'

-Dataset VXSYTSYG (0.19) new variables: [phone]

NCPCAPAB='NOMINAL*CAPABILITY*OF A*CPU'

FXRDONE ='ZHPF DCW*TRANSLATE*SUCCESSFUL'

FXRWRITE='WRITE CHAN*PRESENTED*TO ZHPF DSW'

-Dataset VXSYTSPT (0.24) new variables: 308 364 106 2169

-PROTECTED: MRHDRLEN=364 ONLY 56 DOCUMENTED.

-Dataset VXMTRPRP (1.05) new variables: 12 56 4 433

CALENTMT='VERTICAL*CPU*ENTITLEMENT='

OFFTOPDS='OFFSET TO RCCTOPDS ARRAY='

PFXPOLAR='CURRENT*POLARIZATION='

RCCTOPDI='DISPATCH*VECTOR*BLOCK='

RCCTOPDS='TOPOLOGY*DESCRIPTOR*IDS='

SIZTOPDS='SIZE OF*RCCTOPDS*FIELD='

-Dataset VXMTRMEM (1.07) new variables: 8 168 84 2497

RSACKMB2G='FRAMES*LT 2G*FOR SAD*CRASHKERNEL'

RSACKMA2G='FRAMES*GT 2G*FOR SAD*CRASHKERNEL'

-Dataset VXMTRSCH (1.16) new DATA UNKEPT. 40 132 94 1345

SRXCPPAD and SRXEXUSE decoded but not kept.

-PROTECTED: SRXEXUSE is 8 bytes when PUMAX is 6.

-Dataset VXMTRSSI (1.25) only read SYSPLXNR, in use slots,

but there are SYSPLXNS, configured slots, segments in the

record. All are now input; not-in-use have blank name.

-Dataset VXSCLSTP (2.08) new variables: 4 84 98 3133

SYSFPGRAT='SYSTEM*PAGE*READ IN*RATE*PERSEC'

-Dataset VXSTORSG (3.01) new variables: 300 764 114 121

RSAAGEPC ='TGT SIZE*GLOBAL*AGING LIST*PCT DPA'

RSAAGEFL ='FLAG*BYTE'

RSARSDMX ='SET RESERVED SYSMAX VALUE FRAMES'

RSAAGESZ ='TGT SIZE OF GLOBAL AGING LIST*FRAMES'

RSAAGINC ='FRAMES*ON THE*GLOBAL*AGING*LIST'

RSAEWNDD ='GAL*CHANGED PAGES*MUST BE*WRITTEN'

RSAEWRFO ='GAL*REF ONLY*NOT HAVE*BE WRITTEN'

RSAEWCIF ='GAL*CHANGED*PAGES*BEING*PROCESSED'

RSAEWRIF ='GAL*REF ONLY*PAGES*BEING*PROCESSED'