IBM 5655-DB2 Manuel D’Utilisation
Case Study
149
12.2.1.1 Device Activity Report Analysis
The RMF report analysis is based on the LCU level and Storage Group level
device activity reports. Figure 68 on page 149 shows the extracted summary
lines. The Storage Group level line is the average combination of LCUs 46-48 and
4A-4D activities, as the detail volume level display shows (refer to Figure 49 on
page 127 as an example). The other information to review is:
device activity reports. Figure 68 on page 149 shows the extracted summary
lines. The Storage Group level line is the average combination of LCUs 46-48 and
4A-4D activities, as the detail volume level display shows (refer to Figure 49 on
page 127 as an example). The other information to review is:
• DEVICE ACTIVITY RATE is spread across both LCUs and RVAs inside the
Storage Group, which indicates an allocation controlled environment. The
level of activity is 90.959.
level of activity is 90.959.
• AVG RESP TIME is 31 ms.
• AVG IOSQ TIME is 1 ms, which also leads to a controlled environment.
• AVG PEND TIME is 0.2 ms, not an issue.
• AVG DISC TIME is 7.1 ms, which is quite high but lower than the connect time.
This requires further analysis.
• AVG CONN TIME is 22.7 ms, which indicates heavy transfers. This
information matches the DB2 PM overview of 32 pages, each consisting of
one 4K CI, plus some overhead. The accuracy of this estimate is based on
supposed homogeneous allocation parameters in the same Storage Group.
one 4K CI, plus some overhead. The accuracy of this estimate is based on
supposed homogeneous allocation parameters in the same Storage Group.
From the CONN time and ACTIVITY rate, the path demand is deduced thus:
( ( 90.959 x 22.7) / 1000 ) x 100 = 206.5 % of a path demand
This Storage Group level device activity report is only for one LPAR. Several
LPAR reports should be consolidated (with the spreadsheets) to get the global
Storage Group demand in a data sharing environment. The case study is for only
one LPAR. A more complex situation would be several LPARs spead across
different CPCs, with EMIF shared channels.
LPAR reports should be consolidated (with the spreadsheets) to get the global
Storage Group demand in a data sharing environment. The case study is for only
one LPAR. A more complex situation would be several LPARs spead across
different CPCs, with EMIF shared channels.
Figure 68. Case Study RMF Direct Access Device Activity Report Extract
12.2.1.2 I/O Queuing Activity Report Analysis
Figure 69 on page 150 displays an extract of this report, which enumerates the
CHANNEL PATH hexadecimal identifications for each LCU:
CHANNEL PATH hexadecimal identifications for each LCU:
D I R E C T
A C C E S S
D E V I C E
A C T I V I T Y
OS/390
SYSTEM ID QP02
START 02/12/1999-15.05.39
INTERVAL 000.38.01
REL. 02.06.00
RPT VERSION 2.6.0
END
02/12/1999-15.43.41
CYCLE 1.000 SECONDS
( RMF report extract )
DEVICE
AVG
AVG
AVG
AVG
AVG
AVG
AVG
AVG
%
%
%
AVG
%
%
STORAGE
DEV
DEVICE
VOLUME
LCU
ACTIVITY RESP IOSQ
DPB
CUB
DB
PEND DISC CONN
DEV
DEV
DEV
NUMBER ANY
MT
GROUP
NUM
TYPE
SERIAL
RATE
TIME TIME
DLY
DLY
DLY
TIME TIME TIME
CONN
UTIL
RESV
ALLOC
ALLOC
PEND
LCU
0046
15.778
32
0
0.0
0.0
0.0
0.2
7.1 24.5
0.60
0.78
0.0
15.0
100.0
0.0
LCU
0047
8.752
33
2
0.0
0.0
0.0
0.2
6.7 24.7
0.34
0.43
0.0
14.0
100.0
0.0
LCU
0048
14.145
27
1
0.0
0.0
0.0
0.2
5.4 20.6
0.45
0.57
0.0
11.0
100.0
0.0
LCU
0049
7.923
32
0
0.0
0.0
0.0
0.2
7.0 24.7
0.31
0.39
0.0
19.0
100.0
0.0
LCU
004A
3.721
34
0
0.0
0.0
0.0
0.2
9.3 24.2
0.14
0.20
0.0
3.0
100.0
0.0
LCU
004B
14.431
26
2
0.0
0.0
0.0
0.2
6.3 17.8
0.40
0.54
0.0
4.0
100.0
0.0
LCU
004C
14.540
35
0
0.0
0.0
0.0
0.2
9.1 25.1
0.57
0.78
0.0
5.0
100.0
0.0
LCU
004D
11.668
30
0
0.0
0.0
0.0
0.2
7.6 22.3
0.41
0.55
0.0
7.0
100.0
0.0
LCU
0055
6.963
3
0
0.0
0.0
0.0
0.4
0.1
2.6
0.06
0.06
0.0
299
100.0
0.0
RVA1
SG
90.959
31
1
0.0
0.0
0.0
0.2
7.1 22.7
0.40
0.53
0.0
78.0
100.0
0.0