The INDICATE command is frequently used by z/VM administrators to monitor system status. This Technote discusses interpret output of the INDICATE LOAD.
z/VM has some basic commands such as INDICATE. There are many parameters that can be included. Use the HELP INDICATE command for a basic understanding of the INDICATE command and then press F11 for help regarding each parameter.
If no parameter is specified, INDICATE LOAD is the default option. There are two types of this option: A class G and a class E. Class G users can use INDICATE to display recent contention for system resources and display environment characteristics and measurements of resources used by their virtual machine. The output from a user ID with class E privilege, such as MAINT, OPERATOR, is shown here. The lines are numbered for clarity in the description that follows.
==> IND LOAD
1 AVGPROC-038% 03
2 XSTORE-000021/SEC MIGRATE-0001/SEC
3 MDC READS-000068/SEC WRITES-000001/SEC HIT RATIO-099%
4 STORAGE-015% PAGING-0031/SEC STEAL-000%
5 Q0-00006(00000) DORMANT-00357
6 Q1-00001(00000) E1-00000(00000)
7 Q2-00001(00000) EXPAN-002 E2-00000(00000)
8 Q3-00034(00000) EXPAN-002 E3-00000(00000)
10 PROC 0000-038% PROC 0001-038%
11 PROC 0002-038%
The INDICATE LOAD command gives a snapshot of current system performance. Except for the counts of virtual machines in various queues and the limited list, the values that you see here are a smoothed average over the past four minutes. Areas where z/VM performance analysts tend to focus are:
- AVGPROC, on line 1, gives the overall processor utilization, which is 38 percent in this example. The number following it is the number of online processors, which is three in this example. The individual processor utilization is shown on lines 10 and 11. Look at these to see if they are somewhat balanced. There are cases where an imbalance is okay. This includes low-utilization scenarios or cases where there are not enough ready-to-run virtual processors to keep the physical processors busy.
- Line 2 describes paging to expanded storage. Most z/VM systems on z900 class machines can sustain thousands of paging operations a second without any problems. The MIGRATE rate is the number of pages per second being moved from expanded storage out to paging space on DASD. A healthy system has a MIGRATE rate that is significantly lower than the XSTORE rate. However, there are times that the MIGRATE value spikes for brief periods of time.
- Minidisk cache (MDC) statistics are given on line 3. The effectiveness of MDC can be judged by the combination of the READS rate and the HIT RATIO. If both are high, then a large number of physical I/Os are avoided due to the MDC feature. However, a high HIT RATIO is not good if the READS rate is low.
- Line 4 describes more storage (memory) management. The first value, STORAGE utilization on line 5, is often not helpful in determining problems because its computation is not straightforward. The PAGING rate on the same line is more important, and higher values often impact performance. The STEAL percentage is often misleading. This is basically the percentage of pages taken from guests that z/VM believes are non-dormant. Since some guests have periodic timers going off, they appear to be active to z/VM even when relatively idle. Pages taken from these guests are considered stolen. There are scenarios where a system only has active guests, in which case all pages taken are considered stolen. Keeping this in mind, if a high STEAL value is observed, you need to check the paging rate. If the paging rate is low, then the STEAL value is not important.
- On lines 5 through 8, you also see a series of counters that represent the users in various queues. The z/VM scheduler classifies work into three different classes (1 through 3) and a special class of zero. The Qx and Ex column values represent the virtual machines in the dispatch list and the eligible list. The most important value here to validate is that there are no virtual machines in the Eligible list: E1, E2, E3. This implies that z/VM has stopped dispatching some virtual machines to avoid over committing resources. Do not worry about the values in parentheses.
This material has not been submitted to any formal IBM test and is published AS IS. It has not been the subject of rigorous review. IBM assumes no responsibility for its accuracy or completeness. The use of this information or the implementation of any of these techniques is a client responsibility and depends upon the client's ability to evaluate and integrate them into the client's operational environment.