Database Tables and Transparent Tables
Understand how SAP stores data: transparent tables, pooled tables, cluster tables, primary keys, foreign keys, and table relationships.
Database Tables and Transparent Tables
What You'll Learn
- How SAP stores data in the underlying database
- What transparent tables are and why they matter
- Primary keys and foreign keys in SAP
- Table categories: transparent, pooled, and cluster
- How to view table data with SE16N
Transparent Tables
A transparent table is the most common table type in SAP. It has a 1:1 mapping between the dictionary definition in SE11 and the actual database table — same name, same fields, same structure.
SE11 Definition: ZEMPLOYEES Database Table: ZEMPLOYEES
┌──────────────┬──────────────┐ ┌──────────────┬──────────────┐
│ MANDT (key) │ CLNT(3) │ │ MANDT │ VARCHAR(3) │
│ EMP_ID (key) │ NUMC(10) │ │ EMP_ID │ VARCHAR(10) │
│ NAME │ CHAR(50) │ │ NAME │ VARCHAR(50) │
│ DEPARTMENT │ CHAR(30) │ │ DEPARTMENT │ VARCHAR(30) │
│ SALARY │ CURR(13,2) │ │ SALARY │ DECIMAL(13,2)│
└──────────────┴──────────────┘ └──────────────┴──────────────┘
Dictionary Database
What you define in SE11 is what exists in the database. When you activate a table in SE11, SAP creates (or alters) the corresponding database table automatically.
Primary Keys
Every SAP table has a primary key. The first key field is almost always MANDT (client). The remaining key fields uniquely identify a record within that client.
Table KNA1 (Customer Master):
Primary Key: MANDT + KUNNR
MANDT KUNNR NAME1 LAND1
──────────────────────────────────────────────
100 0000001001 Acme Corp US
100 0000001002 Global Industries DE
200 0000001001 Test Customer US ← Same KUNNR, different client
The combination MANDT + KUNNR is unique. Two rows can have the same KUNNR if they're in different clients (different MANDT values). This is how client isolation works at the database level.
Foreign Keys
Foreign keys define relationships between tables:
VBAK (Sales Order Header) VBAP (Sales Order Items)
┌────────┬────────┬────────┐ ┌────────┬────────┬────────┬────────┐
│ MANDT │ VBELN │ KUNNR │ │ MANDT │ VBELN │ POSNR │ MATNR │
│ 100 │ 100001 │ 001001 │───┐ │ 100 │ 100001 │ 000010 │ MAT001 │
│ 100 │ 100002 │ 001002 │ │ │ 100 │ 100001 │ 000020 │ MAT002 │
└────────┴────────┴────────┘ │ │ 100 │ 100002 │ 000010 │ MAT003 │
│ └────────┴────────┴────────┴────────┘
└──── VBELN links header to items
VBAK.VBELN → VBAP.VBELN — every item belongs to a header. This is a standard parent-child relationship, like orders and order_items in any e-commerce database.
Viewing Table Data with SE16N
Type /nSE16N, enter a table name, and press F8 (Execute):
* SE16N lets you:
* - View all records or filter by specific fields
* - Export data to a spreadsheet
* - Count records matching your criteria
* - Display individual records in detailSE16N is your best debugging friend. When an ABAP program produces wrong results, the first question is always: "What's actually in the table?" SE16N answers that.
Useful SE16N tricks:
- Enter
*in a field to see all values - Enter a range like
001000to001999for number ranges - Use the "Number of Entries" button to count without displaying all rows
- Maximum display is 500 rows by default — change the setting if you need more
Table Categories in S/4HANA
In modern S/4HANA, only transparent tables exist. But in older SAP systems, you'll encounter:
Transparent Table → 1:1 with database table (99% of what you'll work with)
Pooled Table → Multiple SAP tables stored in one database table (legacy)
Cluster Table → Multiple SAP tables stored in one database table, compressed (legacy)
Pooled and cluster tables were used for performance in older databases. S/4HANA on HANA doesn't need them — they're all converted to transparent tables during migration.
Creating Sample Data for Practice
REPORT z_create_sample_data.
* Define table structure (should match your SE11 table ZEMPLOYEES)
DATA: ls_emp TYPE zemployees.
* Insert records
ls_emp-mandt = sy-mandt.
ls_emp-employee_id = '0000000001'.
ls_emp-name = 'Alice Johnson'.
ls_emp-department = 'Engineering'.
ls_emp-salary = 120000.
ls_emp-hire_date = '20200115'.
INSERT zemployees FROM ls_emp.
ls_emp-employee_id = '0000000002'.
ls_emp-name = 'Bob Smith'.
ls_emp-department = 'Marketing'.
ls_emp-salary = 95000.
ls_emp-hire_date = '20190601'.
INSERT zemployees FROM ls_emp.
IF sy-subrc = 0.
COMMIT WORK.
WRITE: / 'Data inserted successfully'.
ELSE.
WRITE: / 'Insert failed — data may already exist'.
ENDIF.We'll cover INSERT, UPDATE, DELETE in detail in Lesson 16. This is just to create test data for the next lesson.
Common Mistakes
- Forgetting that MANDT is part of the key. When you select data, SAP automatically filters by the current client (sy-mandt). But if you're doing cross-client operations or debugging, remember that the same key can exist in multiple clients.
- Using SE16 to modify production data. SE16 can display and (with authorization) modify data directly. This is extremely dangerous in production — changes bypass all application logic, validation, and audit trails. Never modify production data via SE16.
- Not understanding key fields. If you try to INSERT a record with the same primary key as an existing record, it fails (sy-subrc = 4). Use MODIFY instead if you want to insert-or-update.
Key Takeaways
- Transparent tables have a 1:1 mapping between SE11 definition and the database table.
- Primary keys always include MANDT (client) as the first field.
- Foreign keys define relationships between tables (header ↔ items, master ↔ transaction).
- SE16N is your primary tool for viewing table data — essential for debugging.
- In S/4HANA, all tables are transparent. Pooled and cluster tables are legacy concepts.
Next Lesson
You know how tables are defined and structured. In Lesson 15: Open SQL — Reading from the Database, we'll write SELECT statements to query SAP data — the operation you'll perform most often as an ABAP developer.