Hi, I provide training in SAP UI5. Also feel free to checkout my new web page https://saptechblog.com
SAP HANA

Introduction to SAP HANA

What is SAP HANA?

-          In memory database platform that is deployable on premise (inside your company) or on demand (Cloud).

Advantages of HANA:

-          Innovative in-memory RDBMS that makes full use of hardware’s and improves application performance.

-          It reduces the data movement between servers as the data to be processed is directly available in memory next to the CPU. It enables quick processing of data. As all the data is available in the main memory it improves performance by eliminating the read/write operation to data in the disk.

-          Eliminates the need for additional application server layer. A thin application server layer is embedded within the database layer itself where it processes the control logic.

 

SAP HANA can be classified on two perspectives.

1.      In-Memory database

2.      Developer tool

 

SAP HANA In-Memory Database:

HANA runs on multicore CPU processes, with fast communications between the CPU cores and contains terabytes of main memory.

All the data to be processed are always available in the main memory and thus eliminates the read for disk I/O.  However, a persistent database is needed to store/backup the data in case of an unexpected event or power failure.

Advantages of the new RDBMS systems:

1.      Columnar data storage

2.      Parallel processing

3.      Simplifying applications

 

Columnar data storage:

A table can be represented as row order or column order.

Row Order:  It stores the data in the table in the sequence of records. Consecutive memory location will contain data for one completed row.

Column Order: All data of one particular column is stored in consecutive memory location and so on.

clip_image002

SAP HANA is mainly optimized for the column storage of data.

Advantages of using column storage:

1.      Avoid need for the creation of secondary index, as each and every column itself acts as an index. Eliminating the secondary indexes reduces complexity of the table and avoids the need of defining and maintaining metadata.

2.      Better data compression compared to the row store.

3.      Scan and join operation on data are faster. Reason behind this is, in columnar storage the data are stored in a consecutive memory location and are loaded into CPU cache faster, where as in row storage the data of the same column is split across different multiple location and thus reduce the performance due to the cache misses.

 

Parallel processing:

SAP HANA was specifically designed for columnar storage. In columnar storage data are already vertically partitioned.  This helps in processing each column with a different CPU core.  Not only that a column itself can be partitioned into multiple parts and each of which can be assigned to a separate CPU core for processing.

clip_image004

 

Simplifying Applications:

When huge volume of data is being processed then the aggregates (sum, count, max, min etc.) of data are being computed and stored as material aggregates after each write operation or at scheduled times. The traditional application uses these material aggregates to improve the application performance.

With HANA, it is possible to calculate aggregates on large amount of data on the fly as the scanning speed of HANA is several gigabytes per millisecond. With calculation on the fly it eliminates the need for the material aggregates. Unlike material aggregates the data are always up to date as there is no need of updating the material aggregates.

 

About Arun

0 Comments:

Post a Comment

Powered by Blogger.