The CO-PA Accelerator option is included in the new specialised SAP HANA Quick Sizer tool that has been available since the second part of 2014.

Sizing for SAP Suite on HANA is covered in SAP Note 1793345 as well.

 

THE FOLLOWING INFORMATION IS NOW OUT OF DATE

 

The CO-PA Accelerator material is included in the SAP Quick Sizer tool, which is now accessible for SAP HANA.

The QuickSizer for CO-PA Accelerator has very simple operating procedures.

1) Start the QuickSizer

2) Type in the customer number.

3) Specify the project name.

4) Choose SAP In-Memory Computing.

5) Choose HANA Rapid Deployment Options.

6) Find the CO-PA-ACC row element.

7) Enter the estimated compression factor, the number of CO-PA users, and the source data footprint in GB (7 is the default)

8) Press the Results Calculate button.

A screenshot from Step 7…

What does “number of users” mean? The number of named users or concurrent users is not what matters.

rather, SAP Note 1514966, which is extracted here, explains the user notion.

 


To determine the maximum number of users who are active (i.e.

produce any action on the server within the specified time

of one hour) that a HANA server can manage, we have selected

the subsequent strategy.

… we presum that there are three sorts of HANA queries.

(easy, medium, and heavy), which vary in how much CPU they require

they need in terms of resources. “Medium” queries typically use twice

10 times as many resources as “easy” queries, whereas heavy queries use twenty

many times more resources.

Additionally, we presume that there are three categories of HANA users:

(“sporadic,” “normal,” and “expert” categories). The categories of users

are determined by the number of times a query is executed and the mixture of

several sorts of inquiries. Usually, “Sporadic” users

run a single query every hour, 80% “simple” queries, and 20%

11 inquiries are processed each hour by “typical” users using “medium” queries.

run 50% “easy,” 50% “medium,” and 50% “expert” queries.

Users conduct 33 queries every hour, 100% of which are “intensive” queries.


 

 

What is the footprint of the source data, and how is it recorded?

 


The amount of data that HANA must store determines how much memory it needs.

is should be kept in memory, or how much disc space

excluded from the data covered by the respective database tables

corresponding indexes. If the database supports it, take note.

The space required by the uncompressed data must be compressed.


 

The aforementioned SAP note also offers guidance on how to get the uncompressed table sizes.

utilising a saved procedure or script. Depending on the source database, the needed procedure will vary, but one of

Choosing the appropriate ERP tables is one of the script’s inputs.

So how do you choose which CO-PA tables get into the source data footprint’s input?

Using transaction KECM or another quick method, you can quickly determine the source data footprint.

RKE ANALYSE COPA report.

From the menu, select “Transaction Data.”

Select the radio button after checking the box next to “Data Volume of Specific Tables” at the bottom.

“Building in the Foreground.” Implement the report. You can come back to this once you’ve completed the analysis.

before clicking Execute, open the transaction and choose the option button “Display Analysis Data.”

The report output will show record counts and memory use in MB after execution.

for every table of CO-PA. We are only interested in the CE1, CE2, and CE4 for the CO-PA Accelerator.

replicating tables into SAP HANA But in order to use BusinessObjects reporting, other

Depending on the precise reporting needs, tables might need to be copied into SAP HANA.

The potential magnitude of the memory usage numbers is determined by multiplying the number

by the ABAP table length and translating the result to MB. of real records in the table. the actual table, though

Because many CO-PA fields will be blank, the database’s size will be reduced. However, this is a simple task.

a simple and efficient approach to determine CO-worst-case PA’s uncompressed sizing. Simply divide the memory use.

multiplying the numbers by 1024 and entering them into the QuickSizer in GB.

Use transaction SE11 to display the memory consumption figures in order to better comprehend them.

a CO-PA desk.

Then choose Table Width from the Extras menu.

It shows the Width of Table popup window.

The memory consumption figure above can be demonstrated by multiplying the number of records in the

the ABAP table x Length:

When divided by (1024)2, 156,949,206 bytes (69,019 x 2,274) is 150 MB, rounded up.

as seen in KECM.

You will use SLT if your CO-PA Accelerator scenario includes BusinessObject reporting.

to duplicate the related master data and text tables in addition to the CO-PA transaction data.

putting SAP HANA. You may want to include these additional data sources in the source data footprint.

dimensions of the QuickSizer tool’s tables. (However, based on early feedback from customers, the master

Little impact is made by data and text tables on HANA sizing.)

However, you can run the function module to find the CO-PA relation tables.

Transaction SE37’s RKE FILL FIELD TABLE.

Enter the Operating Concern after hitting F8 or selecting the Test/Execute icon.

For costing-based CO-PA, use “1” in the PA TYPE field for field ERKRS.

When you click Execute once more or press F8, the Table results will appear.

Select Details View/Edit from the menu. Next to the 228 Entries under FIELDTAB, view Icon.png.

All of the master data tables indicated under CHECKTABLE must be noted.

Take notice of all the text tables listed under TXTAB as you scroll to the right.

You can use the script described in SAP Note 151496 to obtain the source data footprint for these tables.

Alternatively, you could use the technique to manually calculate the Memory Consumption for each table.

illustrated above, where you might get record counts for each table by using transaction SE16.

As a final method, you can also get the source data footprint by viewing the table sizes from transaction DB02.

However, there are a few issues to be aware of when using DB02. Initially, merely provide the Data size and omit the

In the QuickSizer tool, the ndex size. For the QuickSizer, remember to convert from KB to GB as well.

However, if database compression is enabled on the table, you shouldn’t utilise DB02 as this won’t give

You should enter the ideal (uncompressed) value into the SAP HANA QuickSizer.

Here is an example of the results the QuickSizer programme produces based on 100 users and

250 GB of the original data

CPU class – S (Small)

Total SAPS: 46,200

XS memory category (eXtra Small)

Memory: 73,728 megabytes

GB on HANA disc: 358

Class A SCU

The data listed above should be adequate for you to discuss with a hardware supplier the size of HANA appliance that could be needed to instal the CO-PA Accelerator.

Finally, if your firm anticipates development in the

CO-PA tables because of extended data retention periods, because of the expansion of ERP to new business units, or if

You want to transfer old CO-PA data to SAP HANA.