SAP HANA 2.0 Administration

Mark Mergaerts, Bert Vanstechelman

• Publisher: SAP PRESS
• Year: 2021
• Print ISBN: 9781493221042
• eBook ISBN: 9781493221059
• Edition: 1
• Topic: Computer Science
• Subtopic: Computer Science General
• Index: Computer Science

Novice or expert, this is your one-stop shop for administering SAP HANA 2.0! You'll begin with a deep dive into database and engine architecture. Then explore your key tools: SAP HANA cockpit, SAP HANA Studio, the command-line interface, and more. From there, choose the topics you need. Backup and recovery? High availability? Data tiering? Security? Objects, tables, and transactions? All covered within these pages! From installation to performance tuning, this book has it all. Highlights include: 1) Installation and updates
2) Architecture
3) SAP HANA cockpit
4) SAP HANA Studio
5) Persistency
6) High availability/disaster recovery
7) Scale-out systems
8) Data tiering
9) Objects and tables
10) Security
11) Performance
12) Monitoring

1 SAP HANA Overview

SAP HANA is an in-memory, column-oriented, relational database system capable of handling huge amounts of data. SAP HANA can be used in a variety of scenarios. In this chapter, we give an overview of the SAP HANA capabilities and the different use cases in which SAP HANA brings added value to your business.

SAP HANA is a column-store in-memory database capable of handling huge amounts of memory in real time. SAP HANA allows you to deploy online transaction processing (OLTP) and online analytical processing (OLAP) applications in the same database without the need to duplicate data in aggregates.

SAP HANA is much more than just a database. The software appliance layer includes SAP HANA extended application services, which allows you to develop and run your own applications on top of the database system. There is a full-text processing and search engine capable of handling structured and unstructured data. Some features are delivered as add-ons, such as SAP HANA Predictive Analysis Library (PAL) and SAP HANA Business Function Library (BFL). Data provisioning and data integration are supported using real-time replication tools.

The speed of SAP HANA isn’t only related to its in-memory capabilities. The concept of the column store plays an important role in its performance increase. This a fundamental concept will be discussed in detail in the next section. After the explaining the concept of the column store, we’ll dive into the specific hardware requirements for SAP HANA, after which we continue with the use cases for an in-memory database, data provisioning, advanced analytics, and data integration.

1.1 The Column Store

One specific feature you should certainly be aware of is the distinction between a column store and row store. Combined with the in-memory storage of the database, the column store is a key factor in the tremendous performance that an SAP HANA database is able to deliver.

The traditional concept of a table in a relational database is row-oriented: one row, made up of a series of columns or fields, is followed by the next row, containing these same columns, and so on. This is also the way data is stored in a conventional relational database. While row-based storage is generally a good solution for transactional (OLTP) access, it has a significant drawback in an analytical (OLAP) context, where typically large volumes of data are queried using complex selection criteria. A very simple one-table example will illustrate this: a table contains, among others, the fields CUST (customer), ORDNR (order number), and PURCH_DATE (purchase date). In a row-based table, this looks schematically as shown in Figure 1.1.

Figure 1.1 Row-Based Table

Suppose that an OLAP query wants to retrieve all orders made by customer 00125 with a purchase date in the second quarter of 2020. The database engine will scan the table for matching column values, but in a row-store table, the average distance between two consecutive values for a given column is the average physical length of an entire table row. If, for example, the average row length is 1,000 bytes and the database has a physical data block size of 8 KB, then only eight sets of column values can be read per data block. As a result, the scanning process may have to traverse a huge number of data blocks, not all of which will be present in the data cache in memory, thus causing an expensive disk read. Conventional databases can mitigate this to some extent by creating indexes, but each additional index imposes extra write overhead, and indexes are therefore best restricted to the columns most often used in queries.

There is a neat architectural solution to this problem: what if the data were stored column by column instead of row by row? All customer values would be tightly packed together, and so would the other columns. Large sets of values could then be scanned at immense speed. Adding indexes to get decent performance would become unnecessary. The idea of column-based data storage is not new, and SAP HANA is not the only database to have it; for example, SAP IQ (formerly Sybase IQ) is also a column-based database, but it uses traditional disk storage, not in-memory storage. In the case of SAP HANA, the combination of in-memory storage and use of a column store makes it possible to obtain query results in seconds instead of hours. (Sorry, that sounds like marketing talk—but it’s the truth.)

Column-based storage in SAP HANA goes several steps further than simply organizing the data in columnar fashion. The amount of data for each column is further reduced by using various methods of data compression. The most important of these is dictionary-based encoding, which replaces actual column values (like character strings) with a simple, compact reference. In the end, the column-store counterpart of our example might look like the tables in Figure 1.2 (again, this is an oversimplification).

Figure 1.2 Column-Based Table

The top half of the figure shows the dictionaries for the three columns, where each distinct value is assigned a simple and short identifier. The bottom ha...

Table of contents

• Dear Reader
• Notes on Usage
• Table of Contents
•   Preface
• 1   SAP HANA Overview
• 2   Architecture of the SAP HANA System
• 3   Administration Tools
• 4   Instance Administration
• 5   Database Administration
• 6   System Replication
• 7   Scale-Out Systems and High Availability
• 8   Data Tiering
• 9   Planning and Setting Up the SAP HANA Landscape
• 10   Installation and Updates
• 11   Database Objects
• 12   Working with Tables
• 13   Security
• 14   Performance Analysis
• 15   Monitoring and Troubleshooting
• A   Setting Up Your Trial System
• B   References and Further Reading
• C   The Authors
• Index
• Service Pages
• Legal Notes