In-memory Computing with SAP HANA Reading notes-chapter II: SAP Hana overview

This document is a reading note for in-memory Computing with SAP HANA on Lenovo X6 Systems Chapter II SAP HANA Overview.

The most important part of this chapter is the SAP HANA architecture

SAP HANA Overview

The SAP Hana database is a memory database that combines rows, columns, and objects to take advantage of the parallel processing power of hardware.

SAP HANA Appliance is an integrated hardware and software facility with SAP HANA at its core.

SAP HANA Architecture

The core of Hana is the relational engine (relational database engines), which supports both row-and column-based. Row is suitable for INSERT, update operation, but compression ratio, query performance is inferior to the column type.

Each table can select its own row or column engine, and subsequent conversions can be done.
A row-stored table is loaded at startup. A column-stored table can be loaded at startup or on-demand when accessed.
Row and column storage sharing persistence layer, in-memory data changes are stored to disk through SavePoint, while committed transactions are logged to disk via log, usually a flash drive, as the need to synchronize the write disk.

Relational engines can be used in a variety of ways, including SQL (JDBC/ODBC), MDX (ODBO), and bics (SQL DBC).
Calculation engine allows data to be processed in place without moving data. Hana SQL Script allows some application logic to be executed directly in Hana, which is an extension of SQL, and is similar to Oracle's/pl.

SAP HANA Appliance

The soft and hard integrated solution, the core is Hana, plus some management software.

SAP HANA Delivery Model

Supports two modes, new hardware single-vendor deployment mode and multi-vendor integration (available with existing hardware).

SAP HANA as an appliance

The new hardware deployment mode, is the mainstream deployment mode, is the SAP Certified Server Vendor's integrated deployment solution, must be Intel Xeon CPU, each CPU has a certain memory ratio, gigabit network card, redundancy and high availability.

SAP HANA Tailored Data Center integration

Tailored Data Center integration (TDI) mode, which integrates hardware from multiple vendors

Sizing SAP HANA

For the memory computing system, the capacity of memory (CPU memory ratio) is more important, if the match is too high, waste, if the match is too small, the CPU capacity can not be fully played. Therefore, a suitable CPU core memory ratio is required.

Memory per core ratio for SAP HANA appliances

SAP defines cpu/memory ratios in T-shirt size, such as:

S+ uses the memory of S, and the hardware of M. m+.

SAP Business Suite has low CPU requirements, high memory requirements, and different T-shirt Size

Sizing approach

The sizing of Hana depends on what scenario it is for, see SAP Notes for details:
* Note 1514966-sap HANA 1.0:sizing SAP in-memory Database
* Note 1637145-sap NetWeaver BW on hana:sizing SAP in-memory Database
* Note 1793345-sizing for SAP Suite on HANA
* Note 1872170-suite on HANA memory sizing

For Scale-out BW, see the following SAP Notes:

SAP also offers SAP Quick Sizer online sizing tools.

The results of sizing affect the license.

SAP HANA Software Licensing

The SAP HANA software is available in both editions (Platform and Enterprise Edition) and the SAP HANA software Licensing D Epends on the use case.

In-memory Computing with SAP HANA Reading notes-chapter II: SAP Hana overview