TPDC Manual (G06.24+)

Using the SPAM Performance Data Charts and
Reports
Tandem Performance Data Collector (TPDC) Manual—526357-001
5-60
Disc Process Activity (DPREPT)
DISC PROCESS Activity Report for System \CSO (Release D20)
----------------------------
$Volume LDV PC:BC (headings continued below)
======== === -----
(1) (2) (3)
<---MILLISECONDS/REQUEST---> <--------RESOURCE BUSY (%)-------->
<--PER SEC-->
Rqsts/ <Q/SERVICE(MSECS)> <Disc Process(es)> <Disc Process(es)> <---Discs---> Swap Free
Second ExtrnlIntrnl Total Primry Backup Total Discs Primry Backup Total Primry Mirror I/Os I/Os
------ ------ ------ ----- ------ ------ ----- ----- ------ ------ ----- ------ ------ ---- ----
(4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18)
THIS REPORT SHOWS THE OVERALL ACTIVITY OF (9) The average number of CPU milliseconds
DISC PROCESSES DURING THE MEASUREMENT consumed by the backup Disc Process(es) per
INTERVAL (WINDOW). IT ALSO INCLUDES SOME request for service.
MINIMAL INFORMATION ON THE ASSOCIATED
PHYSICAL DISCS. (10) The sum of (8) and (9).
THE REPORT SHOWS ONLY DISC PROCESSES (11) The average number of DISC milliseconds
THAT RECEIVED AT LEAST 0.10 REQUESTS FOR consumed per request for service. The reported
SERVICE PER SECOND. value takes into account whether the volume is
mirrored or unmirrored and, for mirrored volumes,
attempts to account for serial versus parallel
(1) The disc volume name. writes. If controller path switches occurred
during the measurement window, this latter
(2) The Logical Device Number of the disc volume. adjustment may not be accurate.
(3) The Disc Process' primary and backup CPU (12) The percentage of the measurement window that the
numbers. primary Disc Processes were using their CPU.
(4) The average number of requests for (13) The percentage of the measurement window that the
service received per second. backup Disc Processes were using their CPU.
(5) The average time, in milliseconds, that (14) The sum of (12) and (13).
requests for service spent in the Disc
Process' EXTERNAL queue (i.e. before a (15) The percentage of the measurement window that the
Disc Process read the request). primary disc drive was busy seeking, reading or
writing.
(6) The average time, in milliseconds, that
requests for service spent in the Disc (16) The percentage of the measurement window that the
Process' INTERNAL queue, after having mirror disc drive (if any) was busy seeking,
been read from the external queue; this reading or writing.
time includes both queueing and service
times. (17) The average number of I/O transfers per second
that were carried out on behalf of the Memory
Internal queueing of requests may be Manager processes in the system. These values may
caused by such things as: contention for be greater than the values reported in the CPU
the CPU, contention for Disc Cache report 'swaps per second' column, since a Disc
semaphores, contention for the I/O Channel, Process may perform more than one I/O transfer per
I/O Controller(s) or the physical disc(s), 'swap' operation performed by a Memory Manager
or application-level contention for locked process.
records or files. It may also be caused
by requests that are not expected to (18) The average number of I/O operations per second to
complete rapidly; for example, CONTROL-27 the Disc Free Space table. This table must be
(write notification) requests on a file to accessed and updated whenever space is allocated
which writes, updates or deletes are done or deallocated on the volume.
infrequently.
(7) The sum of (5) and (6). Thus, this is the
average 'response time' experienced by a
request for service.
(8) The average number of CPU milliseconds
consumed by the primary Disc Process(es)
per request for service.