Limited supportBMC provides limited support for this version of the product. As a result, BMC no longer accepts comments in this space. If you encounter problems with the product version or the space, contact BMC Support.BMC recommends upgrading to the latest version of the product. To see documentation for that version, see BMC AMI Ops Monitor for CMF 6.3.

I/O Queuing Activity Report field descriptions

The table below describes in the I/O Queuing Activity Report.

Table 1. Field descriptions for the I/O Queuing Activity Report

Field

Description

IODF SECTION

IODF

two-character hexadecimal suffix for the IODF data set name

CREATED

date and time this IODF data set was created

ACTION

text indicating how this IODF data set was activated

I/O PROCESSORS SECTION

IOP

I/O processor (IOP) numeric identifier; the IOP data section is sorted according ascending IOP numbers

Initiative Queue

for each IOP, there is one initiative queue

ACTIVITY RATE

rate, per second, at which the channel subsystem places requests on the I/O processor’s initiative queue

Note

This rate might be greater than the actual I/O rate because of requeued I/O requests.

AVG Q LNGTH

average number of entries present on the initiative queue

IOP Utilization

% IOP BUSY

ratio of the number of times that the IOP was found busy and the total number of samples taken

I/O START RATE

rate, per second, at which I/O requests are initiated on this IOP

INTERRUPT RATE

rate, per second, at which interrupts are processed on this IOP

% I/O Requests Retried, includes all initial I/O requests and all requests that were retried

ALL

percentage of I/O requests that were retried for any reason

CP BUSY

percentage of I/O requests that were retried because of CHANNEL busy conditions

DP BUSY

percentage of I/O requests that were retried because of DIRECTOR PORT busy conditions

CU BUSY

percentage of I/O requests that were retried because of Control UNIT busy condition

DV BUSY

percentage of I/O requests that were retried because of DEVICE busy conditions

Retries / SSCH, versus the number of initial I/O requests

ALL

ratio of I/O requests that were retried for any reason,

CP BUSY

ratio of I/O requests that were retried because of CHANNEL busy conditions

DP BUSY

ratio of I/O requests that were retried because of DIRECTOR PORT busy conditions

CU BUSY

ratio of I/O requests that were retried because of Control UNIT busy conditions

DV BUSY

ratio of I/O requests that were retried because of DEVICE busy conditions

LOGICAL Control UNITS SECTION

LCU

Logical Control Unit identifier; a four-digit hexadecimal number; an LCU is a logical entity representing one or more physical control units

CNTL UNIT

four-digit hexadecimal number of each physical control unit pertaining to this LCU; an LCU can have up to four physical control units

DCM

This data is available only for LCUs with DCM managed channel paths. Additionally, for all DCM-managed channel paths, the I/O activity rate, the director port contention, and the control unit contention are computed and shown. Note that these values can include managed channels that were only partially online.

MIN

minimum number of DCM-managed channels for this LCU for the reporting period

MAX

maximum number of DCM-managed channels for this LCU for the reporting period

DEF

number of installation-defined managed channels for the LCU

CHAN PATH

two-digit hexadecimal number of each channel path that is assigned to this LCU

An LCU can have up to four channel paths. Four asterisks (****) in this column indicate a summary line for all channel paths connected to this LCU.

PREF PATH

path attribute; displayed for subsystems that support Preferred Pathing (for example, the 1750)

The path attribute values that can appear are preferred (PF), nonpreferred (NP), and not specified (NS).

CHPID TAKEN

rate, per second, of I/O requests handled by each CHPID during the reporting period

If a channel path was offline or moved offline or online during the reporting period, this column could contain the following: OFFLINENOW OFFLINENOW ONLINE

% DP BUSY

percentage of I/O requests that were deferred because a director port was busy; this figure is a measure of director port contention

% CU BUSY

percentage of I/O requests that were deferred because the control unit was busy; this figure is a measure of control unit contention along each channel path

AVG CUB DLY

average number of milliseconds that the I/O request for the channel path waited because the control unit was busy

AVG CMR DLY

average number of milliseconds for a start or resume function needed before the first command of the channel program sent to the device is marked Accepted by the device

Note

CMR—Initial-command-response.

CONTNTN RATE

rate, per second, at which delayed I/O requests are placed on the Control Unit Header (CU-HDR) by the IOP

DELAY Q LENGTH

average number of delayed I/O requests present on the CU-HDR

AVG CSS DLY

average number of milliseconds for the start or resume function needed from the acceptance by the subchannel until the channel subsystem attempts to start the function; this figure is a measure of channel subsystem contention along each LCU

AVG SRVC TIME (MS)

average number of milliseconds it took to complete an I/O request for all devices attached to this LCU that were monitored by the device activity sampler

This field is present only if SMF- type 74-1 records are available.

AVG IOSQ TIME (MS)

average number of milliseconds that I/O requests had to wait before a SSCH instruction was issued to queue the request, for all devices attached to this LCU that were monitored by the device activity sampler

This field is present only if SMF- type 74-1 records are available.

HYPERPAV I/O WAIT

ratio of the number of times an I/O could not start (because no HyperPAV aliases were available) to the number of successful I/Os

HYPERPAV MAX USED

maximum number of concurrently in-use HyperPAV alias devices for the specified LCU during the interval

Note

This field is always zero in reports that are produced from these records:

  • IPM records
  • CPM records that are produced with HPVHWM=NO specified in the IOQ Extractor control statement
  • CPM records that are produced by a PAS started with CXEN=N (CXEN= is a parameter on the PAS PROC; the default is Y)

For more information, see the description of the HPVHWM parameter in the IOQ Extractor control statement.

AVG OPEN EXCH

estimated average number of concurrent active I/O operations

This value is calculated as follows:

  • (CMR+CONN+DISC)/Recording_Interval_Time

where

  • CMR = initial command response time
  • CONN = connect time
  • DISC = disconnect time

DATA XFER CONC

data transfer concurrency

This value is calculated as follows:

  • CONN/Recording_Interval_Time

where

  • CONN = connect time

ALIAS MANAGEMENT GROUP REPORT

AMG

Alias Management Group; a four-digit hexadecimal number;

CNTL UNIT

four-digit hexadecimal number of each physical control unit pertaining to this AMG; an AMG can have up to four physical control units

% DP BUSY

percentage of I/O requests that were deferred because a director port was busy; this figure is a measure of director port contention

% CU BUSY

percentage of I/O requests that were deferred because the control unit was busy; this figure is a measure of control unit contention along each channel path

AVG CUB DLY

average number of milliseconds that the I/O request for the channel path waited because the control unit was busy

AVG CMR DLY

average number of milliseconds for a start or resume function needed before the first command of the channel program sent to the device is marked Accepted by the device

Note

CMR—Initial-command-response.

CONTNTN RATE

rate, per second, at which delayed I/O requests are placed on the Control Unit Header (CU-HDR) by the IOP

DELAY Q LENGTH

average number of delayed I/O requests present on the CU-HDR

AVG CSS DLY

average number of milliseconds for the start or resume function needed from the acceptance by the subchannel until the channel subsystem attempts to start the function; this figure is a measure of channel subsystem contention along each LCU

AVG SRVC TIME (MS)

average number of milliseconds it took to complete an I/O request for all devices attached to this LCU that were monitored by the device activity sampler

This field is present only if SMF- type 74-1 records are available.

AVG IOSQ TIME (MS)

average number of milliseconds that I/O requests had to wait before a SSCH instruction was issued to queue the request, for all devices attached to this LCU that were monitored by the device activity sampler

This field is present only if SMF- type 74-1 records are available.

HYPERPAV I/O WAIT

ratio of the number of times an I/O could not start (because no HyperPAV aliases were available) to the number of successful I/Os

HYPERPAV MAX USED

maximum number of concurrently in-use HyperPAV alias devices for the specified LCU during the interval

Note

This field is always zero in reports that are produced from these records:

  • IPM records
  • CPM records that are produced with HPVHWM=NO specified in the IOQ Extractor control statement
  • CPM records that are produced by a PAS started with CXEN=N (CXEN= is a parameter on the PAS PROC; the default is Y)

For more information, see the description of the HPVHWM parameter in the IOQ Extractor control statement.

AVG OPEN EXCH

estimated average number of concurrent active I/O operations

This value is calculated as follows:

  • (CMR+CONN+DISC)/Recording_Interval_Time

where

  • CMR = initial command response time
  • CONN = connect time
  • DISC = disconnect time

DATA XFER CONC

data transfer concurrency

This value is calculated as follows:

  • CONN/Recording_Interval_Time

where

  • CONN = connect time

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*