SAP HANA Dynamic Tiering Setup
What is Dynamic Tiering?:
The SAP HANA dynamic tiering option is a native big data solution for SAP HANA. The dynamic tiering option adds smart, disk-based extended storage to your SAP HANA database. Dynamic tiering enhances SAP HANA with large volume, warm data management capability.
The dynamic tiering option adds the extended storage service to your SAP HANA system. You use the extended storage service to create the extended storage store and extended tables. Extended tables behave like all other HANA tables, but their data resides in the disk-based extended storage store.
Your application automatically determines which tier to save data to: the SAP HANA in-memory store (the hot store), or extended storage (the warm store). When you use dynamic tiering to place hot data in SAP HANA in-memory tables, and warm data in extended tables, highest value data remains in memory, and cooler less-valuable data is saved to the extended store. This can reduce the size of your in-memory database.
[Credits: Dynamic_Tiering_Option_Master_Guide_en]
Dynamic Tiering Landscape Setup:
Dynamic Tiering feature [SAP HANA Extended Storage feature] is supported in since HANA SP09.
This blog is about My Experience on Dynamic Tiering Setup. This blog will helpful for you to make the Dynamic Tiering setup in your landscape.
I have two Linux Hosts [I named as LinuxHost1, LinuxHost2] with same configuration and same root user and password.
HANA Server will be installed on LinuxHost1 and ES Server will be installed on LinuxHost2.
Both the components cannot be installed on same Linux machine.
Shared File systems:
The following file systems in the two Linux hosts should have shared persistence.
[I am going to Install HANA System with SID: HDT]
- /hana/shared
- /hana/data/HDT
- /hana/log/HDT
- /hana/data_es/HDT
- /hana/log_es/HDT
Installation Parameters:
I install HANS System using batch file. [Batch file is stored in LinuxHost1] [File Format *.cfg]
The following Parameters, especially we have to mention for Dynamic Tiering Setup.
1.addhosts:
We have to mention the details of LinuxHost2 and its role.
<Component_dirs>: Location of <HANA_Option_DynamicTiering_Path>
So these two folder paths have been mentioned in <component_dirs> parameter
3. Other Required Parameters:
Then we have to mention all other required parameters in the batch file (cfg file) like SID, Instance Number, DB_Mode (Single Container), sapadm password, system administrator (<SID>adm) password, system_user_password, user_id, group_id, sapmnt path, data path and log path.
Installation Process:
Once Configuration file is ready, we have to start Installation using hdblcm with configuration file.
During Installation Process, To be Installed Components will be listed.
Components are
- Server (SAP HANA DB)
- ES (Dynamic Tiering).
Select the option all, which will install HANA Server in LinuxHost1 and Dynamic Tiering setup [ES Server] in LinuxHost2.
How to Make Dynamic Tiering is running:
After Installation completed, If we check the system, We can see the Information that Dynamic Tiering is Installed But not running.
Execute the query,
SELECT * from M_SERVICES
Co-ordinater Type for esserver is NONE.
So Dynamic Tiering is Installed But not running. How can we make Dynamic Tiering is running?
Create Extended Storage in LinuxHost2:
To make Dynamic Tiering is running, we have to create Extended Storage in LinuxHost2.
Execute the following SQL Query in HDT System.
CREATE EXTENDED STORAGE AT ‘<full_host_name_of_Linuxhost2>’ size 10 GB ENABLE DELTA
Now check M_Services:
The Data is retrieved from extended storage 🙂
Regards,
Muthu
Great content and explaination.
Question - Can a table be both in memory and extended and just the data is spread between the 2 table types?
Hello Pise, what you are describing will happen in the future, one table could have hot partitions in memory and a warm partition in the extended storage server.
Muthu, the minimum memory requirements to install the extended storage in a vm seems to be 8 GB.
Hi
We're in a Customer. We've installed Dynamic Data Tiering
Service is running
If we create a table as EXTENDED, we get this error message:
create table "TEST"( "CODFIELD" VARCHAR (1) not null,
"DESCFILELD" VARCHAR (50) null,
primary key ("CODFIELD")) USING EXTENDED STORAGE
Error Message: "Could not execute 'create table "TEST"( "CODFIELD" VARCHAR (1) not null, "DESCFILELD" VARCHAR (50) null, ...'
SAP DBTech JDBC: [471]: invalid data source name: __esstore$
Any Idea??
Thanks in advance
If we try to create an EXTENDED STORAGE:
CREATE EXTENDED STORAGE AT 'XXXX' size 10 GB ENABLE DELTA
We have this error message:
Could not execute 'CREATE EXTENDED STORAGE AT 'XXXX' size 10 GB ENABLE DELTA' in 22 ms 832 µs .
SAP DBTech JDBC: [403]: internal error: Extended Storage server already exists on host XXXX
If we create and EXTENDED STORAGE without Host Name (optional), the system uses the first available dynamic tiering host where esserver is running in utility mode.
CREATE EXTENDED STORAGE size 10 GB ENABLE DELTA
We have this message:
Could not execute 'CREATE EXTENDED STORAGE size 10 GB ENABLE DELTA' in 26 ms 850 µs .
SAP DBTech JDBC: [403]: internal error: No esserver found running in utility mode on any host.
If we try to delete this EXTENDED STORAGE
DROP EXTENDED STORAGE
We have this message:
Could not execute 'DROP EXTENDED STORAGE'
SAP DBTech JDBC: [471]: invalid data source name: __esstore$
Any idea??
Hi-- We have a query that once the SAP HANA Dynamic Tiering is configured on the HANA running SAP BW 7.4 , how does the BW objects are marked to be kept in the extended storage and what are those BW objects eligible to be stored in DT.
Request to share some pointers and documents if any.
Regards,
Sharad
Very useful guide, after the installation of the HANA DB and before this guide I had to enable the database as the "worker" (main dynamic tiering host);also noticed that the user which will perform this activity to create the tables need to have the system privilege named: EXTENDED STORAGE ADMIN. Hope this reference help to complement this guide.
Regards
Marcos.