Tandem Capacity Model (TCM) Manual

Significant Columns
13. Txn Name—Modifiable by TCM Administrator
Txn Name is the name of the transactions as shown in the Workload Summary screen of the
Baseline WA model.
14. Type—Not directly modifiable
Type indicates the type of transaction. X indicates an online (OLTP) transaction, B indicates a
batch transaction, and Q indicates a query transaction. When the Scratch Areas appear, the
first Type column shows the source transaction type X, B, or Q, and the adjacent column shows
whether the transaction is being modeled as type X or B.
15. Unlabeled Column—Not directly modifiable
The unlabeled column indicates Aud if the transaction is audited but is left blank otherwise.
A transaction is considered audited if it has a nonzero entry in the TMF process category in
the Apportionment Weights table.
16. TPS (or TPH, TPD, or TPM)—Not directly modifiable
TPS, TPH, TPD or TPM indicates the average number of transactions per second, hour, day,
or month and applies to OLTP transactions (type x) only. The TPS is computed from the number
of transactions and the size of the sample window. When the Scratch Areas are visible, the
first TPS column shows the throughput in transactions per second, and the second shows
throughput in the units you select (for example, transactions per hour).
17. #Txns per Task—Not directly modifiable
#Txns per Task applies to batch transactions only. It is the number of subtransactions within
the batch task or transaction itself.
18. CPU Sec—Not directly modifiable
CPU Sec is the apportioned number of CPU seconds (adjusted to include interrupt seconds)
consumed per transaction as shown in the Workload Summary screen of the Baseline WA
model. The Average Txn row contains the (weighted) average of CPU seconds for each
transaction.
19. Disk Sec—Not directly modifiable
The apportioned number of disk read and disk write seconds consumed per transaction as
shown in the Workload Summary screen of the Baseline WA model. The Average Txn column
contains the (weighted) average of disk seconds or each transaction. When you select Display
Scratch Areas from the Display tab, this section expands to show more detailed information.
Consumption Model Screen
The Consumption model calculates the number of CPUs and disks required to achieve acceptable
performance as defined by the capacity planner. The model also estimates host response times,
utilizations, and maximum throughputs.
In the Consumption model, you define acceptable performance by defining the rules the model
should use in developing its planned, or estimated, performance and the requirements of the system.
To do this, specify constraints for items such as host OLTP response time, device utilizations, and
the number of CPUs, disks, or both. Within these constraints, the Consumption model determines
the number of CPUs and disks required and the expected performance characteristics of the
application under study.
When one or more of the constraints is not achievable, a warning message appears with an
asterisk (*) displayed next to the unachievable constraint.
You can also specify planned transaction throughputs that are different from the actual throughputs.
In addition, super users can change the CPU or disk seconds per transaction in the Plan column
to investigate the effect of possible changes in the processing requirements of a given transaction.
140 The Performance Model