IBM® System Storage™ Architecture

and Configuration Guide

for
SAP® HANA™
Tailored Datacenter Integration
location of this document:

https://www.ibm.com/support/pages/node/6355415

[email protected]

Version 2.30 – Sep. 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

Contents
Contents................................................................................................................... 2
1 Preface............................................................................................................. 4
2 IBM storage architecture for SAP HANA TDI .................................................... 4
3 The central theme ............................................................................................ 5
4 Requirements .................................................................................................. 5
5 Installation of SUSE SLES or Red Hat RHEL ..................................................... 6
6 Setup of shared access for the /hana/shared/ directory .................................. 6
7 IBM System Storage sizing & configuration ...................................................... 6
8 Linux OS setup & tuning ................................................................................. 12
9 SAN zoning and volume mapping ................................................................... 14
10 Setup of file systems for DATA and LOG......................................................... 15
11 VMware .......................................................................................................... 16
12 Setup of global.ini .......................................................................................... 16
13 fileio parameter ............................................................................................. 18
14 SAP HANA High Availability setup with IBM System Storage ......................... 18
15 SAP HANA Disaster Recovery setup with IBM System Storage ...................... 19
16 SAP HANA backup with IBM Spectrum Protect .............................................. 19
17 Change History ............................................................................................... 20
18 Resources ...................................................................................................... 21
19 Trademarks .................................................................................................... 22
20 Disclaimers .................................................................................................... 23

Page - 2 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

Page - 3 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

1 Preface

This paper is intended as an architecture and configuration guide to setup the IBM®
System Storage™ for the SAP® HANA™ tailored datacenter integration (SAP HANA TDI)
within a SAN environment.
This document has been written for IT technical specialists and architects, with ad-
vanced skill levels on SUSE® Linux™ Enterprise Server or Red Hat LINUX® and IBM Sys-
tem Storage, with a focus on architecting and setting up an IBM System Storage envi-
ronment for SAP HANA TDI. The SAP HANA TDI allows the SAP customer to use exter-
nal storage to attach to the SAP HANA server.
This document provides the necessary information to select, verify and connect IBM
System Storage to the SAP HANA server through a SAN (fiber channel) attached storage
network. The recommendations in this guideline apply to both single node and scale
out configurations, as well as Intel® or IBM POWER® based SAP HANA systems.
For a list of all storage systems certified for SAP HANA production please visit:
https://www.sap.com/dmc/exp/2014-09-02-hana-hardware/enEN/#/solutions?filters=v:deCerti-
fied;storage;ve:6

2 IBM storage architecture for SAP HANA TDI

The IBM System Storage needs to be connected through a SAN network with the SAP
HANA server. IBM recommends – but not requires – the use of the IBM Spectrum Vir-
tualize (SAN Volume Controller, SVC) as virtualization layer, enabling to centralize the
management of the storage systems, as well as to provide easily high availability and
disaster recovery functionality, as well as IBM FlashCopy backup and tiered storage
pooling. For smaller HANA TDI deployments you can also use the IBM FlashSystem,
which is based on the same functions of SVC, and inherits its advantages described
above.

IBM Storage Systems certified for SAP HANA TDI production are:
• IBM Spectrum Virtualize (SAN Volume Controller, IBM FlashSystem)

Page - 4 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

• IBM DS8900F Family

For non-prod SAP HANA systems, any storage system can be used.

3 The central theme

Below the generic workflow describes how to setup SAP HANA TDI with IBM storage:

I. Verify all server & storage components and firmware requirements given by
SAP are met
II. Install SLES or Red Hat operating system on all SAP HANA nodes
III. If required: setup NFS services for the SAP HANA shared directory
IV. Setup SAN storage systems
V. Create SAN zone for SAP HANA environment
VI. On the storage systems map the storage volumes (LUNs)
to all target SAP HANA nodes. Volumes must be attached via the SCSI proto-
col. NVMe is not supported by SAP HANA.
VII. Perform Linux setup and tuning, e.g. multipath, IO scheduler, ...
VIII. On each LINUX host create the file systems for SAP HANA data, log, shared
IX. Setup of /hana/shared/global.ini for HANA scale out or HANA node failover
X. Install and configure the SAP HANA software
XI. Optional: HA, DR, and Backup considerations

4 Requirements

Before starting to deploy SAP HANA TDI, ensure that the following key items have are
in place:
• A SAN environment is required to attach IBM System Storage through fiber chan-
nel with the SAP HANA TDI nodes.
• SAP requires a storage validation for SAP HANA TDI according to their KPI (key
Performance Indicators) – please contact SAP for further details.
• The SAP HANA nodes must be sized and configured according to SAPs specifica-
tion (Bill of Material for SAP HANA node), in addition each SAP HANA TDI system
needs to have min 4 FC ports, recommended are 8 FC ports.

Page - 5 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

Naming: The term “node” is use for a single HANA instance: if for instance 4 HANA in-
stances are deployed on a single IBM POWER server, or 4 HANA instances are deployed
into four single LPAR, then this reflects 4 nodes.

5 Installation of SUSE SLES or Red Hat RHEL

Install the Linux operating System accordingly to the SAP PAM for the chosen SAP
HANA version, see SAP note https://launchpad.support.sap.com/#/notes/2235581

6 Setup of shared access for the /hana/shared/ directory

SAP HANA scale-out and HANA HA node fail-over require that the directory
/hana/shared/ must be accessible from all nodes, else this directory can be setup as
local file system from type xfs.
The shared access can be achieved most easily through NFS. If such setup does not
already exist, one of the following options can be used to implement it:
Highly Available NFS service with DRBD and Pacemaker with SUSE Linux Enterprise
High Availability Extension.
Setup these systems according to the appropriate implementation guides and make the
shared NFS file system available on all HANA nodes, including the spare nodes.
Setup the NFS Server by adding the following line to file /etc/exports:
/hana/shared node1(fsid=0,crossmnt,rw,no_root_squash,sync,no_subtree_check)
node2(...) node3(...)

To configure the HANA nodes as NFS clients, add the following line to file /etc/fstab on
each HANA node:
nfsserver:/hana/shared /hana/shared nfs4 rw,soft,intr,rsize=8192,wsize=8192 0 0

7 IBM System Storage sizing & configuration

This section covers the evaluated storage options.

Page - 6 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

7.1 Recommended IBM System Storage combinations

To achieve the required storage performance according to SAP KPIs (key performance
indicators) for SAP HANA TDI, one of the following configurations listed should be cho-
sen – nevertheless all IBM systems listed can be used as stand-alone system as well,
used for DATA and LOG.
Because IBM SAN Volume Controller, and for smaller configurations IBM FlashSystem,
provisions volumes (LUNs) as a single volume type and single storage system, it will
ease the administration of storage pool tiers with Flash, SSD, and HDD types; and op-
tional tasks like setup of storage high availability, FlashCopy base backup, storage-
based mirroring, or storage maintenance.

IBM recommends the use of FlashCore Modules (FCM) or NVMe flash drives for produc-
tion workload. From expansion modules, SAS flash drives should be used.

SAS 10k or near line disk drives are suitable for archive data, or for backup as last-
storage-pool.

7.2 NSE Considerations

SAP HANA Native Storage Extension (NSE) is an SAP HANA feature to keep so called
warm data on disk instead of loading it into memory. NSE enabled data is loaded on
demand only.

NSE requires to store the data on fast storage with low IO latencies. Therefore, the stor-
age pool providing /hana/data should be formed of the fastest available drives of the
storage system. For FlashSystem, we recommend using FlashCore Modules (FCM) or
even better Storage Class Memory (SCM).

To avoid any movement of such data to lower storage tiers, ensure also that IBM Stor-
age Easy Tier is disabled on the storage system.

For more information about using NSE with IBM FlashSystem, please refer to
https://www.ibm.com/support/pages/ibm-flashsystem-sap-hana-native-storage-ex-
tension

7.3 IBM System Storage sizing

This chapter provides some general sizing guidelines.

Page - 7 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

All components of the SAN infrastructure must be configured to use a minimum 8GB
link speed with 4 links. All listed storage configurations apply to active SAP HANA TDI
nodes only; no additional storage configuration is required for any stand-by system. If
SVC stretched cluster needs to be configured, all listed storage configuration must be
doubled.
Maximal number of HANA production systems support per IBM System Storage:

IBM System Storage # HANA Prod

IBM FlashSystem 5015/(5035 + 5045)/5200 4 / 18* / 72*
IBM FlashSystem 7300(H) 924 132*
IBM FlashSystem 9500(R) 164**
DS8910F / DS8950F / DS8980F 84 / 150 / 150
SVC 2145/2147 SV3, one / four node pairs 48 / 192
SVC 2145/2147 SA2, one / four node pairs 25 / 100
* with max. supported IOGroups
** IBM FlashSystem 9500 supports 2 IOGroups but doubles the performance of its pre-
decessor, the 4 IOGroup capable IBM FlashSystem 9200

Page - 8 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

7.4 Flash Core Modules (FCM)

During lifetime of all kind of flash technology the capacity utilization should not exceed
80%. If the utilization approaches this limit, additional capacity of the same kind and
device capacity need to be added. For capacity planning and sizing this should be con-
sidered seriously.

7.5 IBM storage performance sizing for HANA

This section provides sizing guideline how to size IBM storage for HANA.
If not stated otherwise (and if applicable), use distributed RAID 6.

The statement “Supports up to xx HANA systems” refer to SAP production systems, in

case of non-prod systems, like Dev/Test/Sand much more HANA system can be used.

IBM SAN Volume Controller

▪ SV3: Each IO group (two nodes) can handle 48 HANA nodes, a four IO-group cluster is able to
handle 192 HANA systems
▪ SA2: Each IO group (two nodes) can handle 25 HANA nodes, a four IO-group cluster is able to
handle 100 HANA systems
▪ Use Pools (MDiskGroups) with at least 8 Backend LUNs (MDisks) for reasonable workload bal-
ancing

IBM FlashSystem 5015 / 5035 / 5045

12 SSD for every 6 HANA systems

IBM FlashSystem 5200

▪ 12 NVMe FCM for every 18 HANA systems

IBM FlashSystem 5200 / 7300

▪ 12 NVMe FCM for every 18 HANA systems

IBM FlashSystem 9500

▪ Supports up to 82 HANA production nodes
▪ It is recommended to use full equipped units only, with 48 NVMe Flash Core Modules.

IBM DS8000 family

▪ 16 Flash Cards for every 8 nodes

Page - 9 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

7.6 Sizing tool MyFlash4HANA

IBM provides a tool to size in IBM storage systems for a given/planed HANA environ-
ment. This tool is available only to IBM and IBM Business Partner.
Please be so kind and ask your IBM sales representative or IBM Business Partner for a
detailed IBM Storage sizing for SAP HANA.

7.7 Definitions of storage pools and volumes

Even though SAP HANA Data and Logs have different I/O characteristics, both work-
loads can be placed on the same SSD / Flash storage pool since Flash technology can
handle any type of workload. With DRAID 6 there is no need to place DATA and LOGs in
two different pools to protect against drive failure.

For all storage systems:

• Create 4 volumes for HANA DATA and 4 volumes for HANA LOG,
for each SAP HANA node, recommendation: thin provisioned
• Switch on Volume / VDISK caching

SVC specific:

• Create at least 8 backend LUNs (MDisks). Increase the number (preferably by

four) in case the system falls short in terms of capacity or performance.
Backend LUNs should always be full provisioned volumes.

7.8 Advanced data reduction features

Implementing storage capacity saving strategies is one of the most wanted features
when configuring new storage for SAP HANA. IBM FlashSystem and IBM SAN Volume
Controller provide a rich bundle of data reduction features which are listed in detail
below. For more detailed information about IBM FlashSystem Capacity Savings with
SAP HANA, please refer to https://www.ibm.com/support/pages/node/6436843

7.8.1 Compression
IBM FlashSystem offers Real time Compression (RtC) and Data Reduction Pools (DRP)
as features to enable data compression, either on VDisk or on Pool level. Because the
data of the HANA database are “just” integers, the expected compression ratio is
slightly small. Using VDisk or Storage Pool based compression is not recommended for
production workloads.
Page - 10 © Copyright IBM Corporation 2023
 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

7.8.2 Deduplication (DRP only)

The IBM storage systems provide the option to deduplicate data on Storage Pool level.
Because the HANA database does internally deduplicate it’s data, the additional capac-
ity saving when using DRP is relatively small compared to other workloads. Using Dedu-
plication is not recommended for production workloads.

7.8.3 IBM Flash Core Modules inline compression

The IBM Flash Core Modules, available for IBM FlashSystem, provides HW bases com-
pression. Inline compression is transparent and has generally no performance impact.
It is a built-in feature of FCM and it is always enabled.

7.9 Layout of storage pools and storage volumes

The following diagrams show the relationship between Linux file system, LINUX LV, VG,
and PV, and the Virtual Volumes (VD, VDISK) and Managed Disks (MD, MDISK,
RAIDs/Arrays) of a Storage Systems.

7.10 IBM SVC stretched cluster configuration

Page - 11 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

IBM SVC stretched cluster

configuration with VDISK mir-
ror is the standard implemen-
tation to gain best high avail-
ability (HA) solution for stor-
age. The following picture il-
lustrates the relationship be-
tween LINUX physical vol-
umes with IBM Virtual and
IBM Managed Disks.

More details can be found in the IBM Redbook " Implementing the IBM System Storage
SAN Volume Controller with IBM Spectrum Virtualize":
https://www.redbooks.ibm.com/abstracts/sg248465.html?Open

7.11 Competing storage utilization

Every IBM storage system can be shared between SAP HANA (production) and any
other SAP or non-SAP (production) workload.

To do a proper sizing, the entire storage infrastructure needs be evaluated, including

the configurations of RAID controller, number of HDD, SSD, or Flash Modules, number
and type of HBA and FC ports/links.

8 Linux OS setup & tuning

General guidelines about required OS setup and tuning provided by SAP can be found
here:
https://launchpad.support.sap.com/#/notes/2235581

Page - 12 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

8.1 Linux host attachment requirements

The Linux operating system default values for SAN attached SCSI devices need some
adjustment prior setting up the host for SAP HANA. Please refer to
https://www.ibm.com/docs/en/flashsystem-7x00/8.4.0?topic=htrlos-attachment-re-
quirements-hosts-that-are-running-linux-operating-system in order to configure the
required SCSI and udev timeouts correctly.

SAP HANA does not support NVMe host attachments, so make sure all LUNs are at-
tached using the SCSI protocol. Please note that this restriction affects host attach-
ments only.

8.2 Linux multipath configuration

IBM recommends specific multipath configurations for IBM Storage Systems. This doc-
umentation can be found in the IBM information web page by using the search function.
For example, the multipath configuration for IBM FlashSystem 7200 is available here:
https://www.ibm.com/docs/en/flashsystem-7x00/8.4.0?topic=system-settings-
linux-hosts

It’s recommended to follow the most recent host attachment guidelines!

8.2.1 Using alias for UUIDs in multipath.conf

The alias setting overrules the user_friendly_names setting.

Using alias makes managing and identifying volumes easier. If you want to use alias,
please make sure that in the case of a multi-node environment, you fully understand
the requirement and handling of consistent multipath device names across all nodes:
The multipath.conf must be identical on all nodes (LINUX systems) belonging to the
HANA instance.
multipaths {
multipath {
wwid 36006048000028350131253594d303030
alias HANA_ANA_DATA_1
}
multipath {
wwid 36006048000028350131253594d303041
alias HANA_ANA_LOG_1
}

Page - 13 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

8.3 IO device tuning

The following tunable parameters can be considered in case I/O performance issues
have been detected. Do not change these in advance. Consult a performance specialist
to find a vital combination:

• Increase /sys/block/<device>/queue/nr_requests if the default (128) results in

blocked I/O submission. This will indirectly help to optimize the blocking inside
SAP HANA.
• Increase 'rr_min_io_rq ' to the power of 2 (4, 8, 16, 32) in multipath.conf
• Increase the queue depth of devices, e.g.
echo 64 > cat /sys/bus/scsi/devices/<device>/queue_depth

8.4 IO scheduler (kernel parameter elevator)

The default scheduler (cfq) isn’t optimal in terms of IO performance for multiple (par-
allel) streaming read processes, therefore it is not recommended for multipath setups.
The NOOP scheduler is recommended for setups with devices that do I/O scheduling
themselves, such as intelligent storage or in multi-pathing environments. The DEAD-
LINE scheduler can provide a superior throughput over the CFQ I/O scheduler in cases
where several threads do reads and writes, and fairness is not an issue. An example for
such a use case are parallel readers from a SAN attached disk via multipathing. Recent
Linux kernels have implemented a true multiqueue I/O scheduling (“blk_mq”). If this
is being used the correct IO scheduler is NONE. Anyway, the choice of the IO scheduler
depends on the installed Linux distribution and it might also differ between several ver-
sions of the same product.

Please ask your Linux vendor for configuring the best IO scheduler.

9 SAN zoning and volume mapping

Apply standard, best practices zoning within the SAN.

If HANA High Availability node-fail over or HANA scale-out is used, then, and only then,
it is required to map all Data and Log volumes to all SAP Hana Nodes.

Page - 14 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

10 Setup of file systems for DATA and LOG

Create the DATA, LOG, and SHARED file system needed by SAP HANA according to the
SAP HANA Storage Requirements documentation from SAP.

• Use the xfs file system type with a 4KB block size for the DATA and LOG file
systems
• To use LVM 2 use this schema:
vgcreate → lvreate → mkfs.xfs
• Use telling names for volume groups and logical volumes.
• Do not use the hyphen (minus) character ‘-‘ as part of the name, because the
LINUX device mapper uses this as a separator between VG and LV name when
it creates device names for the LV.
• Create Volume Groups and Logical Volumes for DATA and LOG, for each HANA
mount point (as listed in global.ini) create one Volume Group with exactly one
Logical Volume.

It has turned out that these settings are most beneficial for SAP HANA workload:

DATA
1. Use 4 PV for Volume Group DATA
pvcreate –dataalignment 1M /dev/mapper/WWID_1 /dev/mapper/WWID_2…
2. Create a volume group with an extent size of 1MB and an alignment size of 1MB
vgcreate -s 1M --dataalignment 1M VG_ANA_DATA_1 /dev/mapper/WWID_1 \ /dev/map-
per/WWID_2 ...
3. Create a logical volume with a stripe size of 256KB
lvcreate -i 4 -I 256K -l 100%VG -n LV_ANA_DATA_1 VG_ANA_DATA_1
4. Create the file system
mkfs.xfs -b size=4096 -s size=4096 /dev/mapper/VG_ANA_DATA_1-LV_ANA_DATA_1
5. Mount the file system
mount /dev/mapper/VG_ANA_DATA_1-LV_ANA_DATA_1 /hana/data/ANA

LOG:

1. Use 4 PV for Volume Group LOG:

pvcreate –dataalignment 1M /dev/mapper/WWID_1 /dev/mapper/WWID_2…
2. Create a volume group with an alignment size of 1MB
vgcreate -s 1M --dataalignment 1M VG_ANA_LOG_1 /dev/mapper/WWID_1 ...
3. Create a striped logical volume with a stripe size of 256KB
lvcreate -i 4 -I 256K -l 100%VG -n LV_ANA_LOG_1 VG_ANA_LOG_1

Page - 15 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

4. Create the file system

mkfs.xfs -b size=4096 -s size=4096 /dev/mapper/VG_ANA_LOG_1-LV_LOG_1
5. Mount the file system
mount /dev/mapper/VG_ANA_LOG_1-LV_ANA_LOG_1 /hana/log/ANA

SHARED (local, non-failover):

1. Use 1 PV for Volume Group SHARED
pvcreate –dataalignment 1M /dev/mapper/WWID_1
2. Create a volume group:
vgcreate -s 1M --dataalignment 1M VG_SHARED /dev/mapper/WWID_1
3. Create a logical volume (without striping)
lvcreate -l 100%VG -n LV_SHARED VG_SHARED
4. Create the file system
mkfs.xfs -b /dev/mapper/VG_SHARED-LV_SHARED
5. Mount the file system
mount /dev/mapper/VG_SHARED-LV_SHARED /hana/shared

If HANA scale-out or HA node fail-over is used, do not add the DATA & LOG file sys-
tem to /etc/fstab, mounting will be done by SAP HANA.
Add the /hana/shared file system always to /etc/fstab.

11 VMware

The same setup applies like for LINUX native installation as described in chapter 10;
the only deviation is that VMware™ ESX© server controls the SAN multipath, not the
LINUX OS. The volumes are typically accessible under the path:
/dev/sd... instead of /dev/mapper/...

Please have a look in the following IBM Redpaper how to setup VMware VVols with IBM
storage:

https://www.redbooks.ibm.com/redpapers/pdfs/redp5321.pdf

12 Setup of global.ini

A dedicated global.ini file (located in /hana/shared/...) is only needed if you want to use
the node fail-over functionality provided by SAP HANA software and for HANA scale-
out; else the HANA installer will create a global.ini file.

Page - 16 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

12.1 Setup for recent versions of HANA 2.0

SAP HANA 2 supports since SPS 05 Rev 53 SCSI persistent reservations on multipath
level. Usually, SCSI reservations are bound to single paths, which makes it difficult to
add SCSI paths to existing volumes while there are SCSI reservations registered. Us-
ing the “mpathpersist” utility allows to deal with this fact.
We recommend using the hdb.ha.fcClientLVMMpath driver, as it gives you a more
powerful protection against double failures (e.g. path failure plus node failover).
Enter the Logical Volume names as shown in directory /dev/mapper into global.ini :

[communication]
listeninterface = .global

[persistence]
basepath_datavolumes = /hana/data/ANA
basepath_logvolumes = /hana/log/ANA

[storage]
ha_provider = hdb_ha.fcClientLVMMpath
partition_*_*__prtype = 5
partition_1_data__lvmname = VG_ANA_DATA_1-LV_ANA_DATA_1
partition_1_log__lvmname = VG_ANA_LOG_1-LV_ANA_LOG_1
partition_2_data__lvmname = VG_ANA_DATA_2-LV_ANA_DATA_2
partition_2_log__lvmname = VG_ANA_LOG_2-LV_ANA_LOG_2

The mpathpersist utility requires predefined SCSI reservation keys dedicated to each
host. The multipath.conf file needs to be modified for using those predefined keys by
adding the reservation_key parameter and an individual value to the defaults section:

defaults {
find_multipaths yes
user_friendly_names yes
reservation_key e4e4063da487427d
}

For more information, please refer to the SAP HANA™ Fiber Channel Storage
Connector Admin Guide provided by SAP.

Page - 17 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

12.2 Setup for HANA 2 < SPS 05 Rev 53

Older HANA 2 revisions do not work with the mpathpersist utility. Therefore, the former
LVM driver must be used. Please note that using this driver is still supported by recent
HANA 2 revisions, it’s not mandatory to update the storage connector after updating
your HANA DB. This is how a typical global.ini file using the standard LVM storage con-
nector looks like:

[communication]
listeninterface = .global

[persistence]
basepath_datavolumes = /hana/data/ANA
basepath_logvolumes = /hana/log/ANA

[storage]
ha_provider = hdb_ha.fcClientLVM
partition_*_*__prtype = 5
partition_1_data__lvmname = VG_ANA_DATA_1-LV_ANA_DATA_1
partition_1_log__lvmname = VG_ANA_LOG_1-LV_ANA_LOG_1
partition_2_data__lvmname = VG_ANA_DATA_2-LV_ANA_DATA_2
partition_2_log__lvmname = VG_ANA_LOG_2-LV_ANA_LOG_2

13 fileio parameter

IBM storage systems do not require a fileio setting (SAP note 2399079) for HANA 2.0.
In case you experience performance issues with the SAP tool HCMT use these settings:
async_write_submit_active : on
async_write_submit_blocks : all
async_read_submit : on

14 SAP HANA High Availability setup with IBM System Storage

The SAP HANA system gains high availability through an N+1 concept – one or more
server act as standby SAP HANA node – with dedicated storage for each active node. If
one active SAP HANA node fails, the SAP HANA cluster software initiates a failover to
the standby node, and the standby node will mount the data and log from the failed
node.
For more details, please read the SAP document SAP HANA – High Availability.

Page - 18 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

To improve storage availability, IBM recommends to setup the storage environment as

SVC stretched cluster with symmetric VDISK mirroring.

15 SAP HANA Disaster Recovery setup with IBM System Storage

The disaster recovery capabilities of SAP HANA are documented on SAP HANA – High
Availability.

If IBM Global Mirror (asynchronous mirroring) is used for storage replication for SAP
HANA TDI, all DATA and all LOG volumes of all nodes from one HANA system must be
included in one storage Consistency Group.
That implies (when using native attached storage) that both – Data and Log – must be
provided by the same storage unit.

16 SAP HANA backup with IBM Spectrum Protect

The product IBM Spectrum Protect for Enterprise Resource Planning includes support to
backup SAP HANA online.

Please see current product information how to install, configure, and run this integrated
solution:

https://www.ibm.com/docs/en/spferp/8.1.11

Page - 19 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

17 Change History

V1 Initial version
V1.2 Update of chapter 10 to clarify files system setup
V1.3 Included new Storwize V5000
V1.4 Specified NFS section to be used only for /hana/shared/
V1.5 Minor updates – add some clarifications
V1.6 Add setup of multipath and global.ini
V1.7 Rework of document
V1.8 Added LINUX LVM2 & list of certified IBM storage systems
V2.1 Added IBM DS8000 system as certified for SAP HANA TDI
V2.2 Added IBM FlashSystem 900 & V9000
V2.3 Changed sizing rules, added recommendation for multipath and VG/LV settings.
V2.5 Changed /etc/multipath setting
V2.6 Added fileio parameter for hdbparam
V2.7 Added A9000R & DS8880; info on LINUX LVM settings for HANA LOG
V2.8 Added chapter on Storage RtC, changes in storage sizing
V2.9 Added paragraph about VMware VVols
V2.10 Updated multipath section
V2.11 Update default disk type to RI SSD.
V2.12 Storwize: New sizing, 4 PV for LOG VG
V2.13 Add SVC 2145/2147 SV1
V2.14 Add FS9100, updated sections 7.7 and 7.9,
jjjj
V2.15 Add V7000G3, minor update to multipath
V2.16 Add V5000E, adding FCM and DRP
V2.17 Minor changes in supported storage systems
V2.18 Minor changes in supported storage systems
V2.19 Adding new IBM FlashSystem models, update DRP section
V2.20 Updating max. numbers of supported systems for FlashSystem, changed DMMP config
V2.21 Minor change: update contact email address
V2.22 Adding NSE section
V2.23 discarded
V2.24 Updating IBM FlashSystem models, major changes in Linux OS setup & tuning section.
V2.25 Minor changes in supported storage systems
V2.26 Added missing content to § 7.7
V2.27 Added max. numbers of supported Systems for IBM SAN Volume Controller SV1 and SA2
V2.28 Adding new IBM Spectrum Virtualize based models
V2.29 Adding section for SCSI multipath reservation configuration, new DS8900 max. sup. DBs
V2.30 Adding updated IBM FlashSystem model 5045

Page - 20 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

18 Resources

For any product or documentation provided by SAP please contact SAP.

For any product or documentation provided by SUSE please contact SUSE.

For any product or documentation provided by Red Hat please contact RedHat.
For questions about products or documentation provided by IBM regarding SAP send
an email to:
[email protected]

For questions regarding this guide please send an email to:

[email protected]

SAP HANA on IBM Power Systems and IBM Storage:

SAP HANA on IBM Power Systems and IBM System Storage

IBM FlashSystem product manuals:

IBM FlashSystem 5200
IBM FlashSystem 7300
IBM FlashSystem 9200

IBM Spectrum Protect for ERP:

IBM Spectrum Protect for Enterprise Resource Planning Data Protection for SAP and Oracle (V8.1)

SAP documentation:
SAP HANA FAQ
SAP HANA™ Fiber Channel Storage Connector Admin Guide

SUSE documentation:
https://documentation.suse.com/

Red Hat documentation:

https://access.redhat.com/documentation/en-us/

Page - 21 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

19 Trademarks

IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of Interna-
tional Business Machines Corporation in the United States, other countries, or both.
These and other IBM trademarked terms are marked on their first occurrence in this
information with the appropriate symbol (® or ™), indicating US registered or common
law trademarks owned by IBM at the time this information was published. Such trade-
marks may also be registered or common law trademarks in other countries. A current
list of IBM trademarks is available on the Web at
https://www.ibm.com/legal/copytrade

The following terms are Trademarks of the International Business Machines Corpora-
tion in the United States, other countries, or both:

Spectrum Storage, System Storage, FlashSystem, Spectrum Protect, Spectrum Virtual-

ize, Spectrum Accelerate, Spectrum Scale, GPFS, IBM, POWER, Redbooks.

The following terms are trademarks of other companies:

SAP, R/3, ABAP, SAP NetWeaver, SAP HANA, the Business Objects logo, BusinessOb-
jects, Crystal Reports, Crystal Decisions, Web Intelligence as well as their respective
logos are trademarks or registered trademarks of SAP SE in Germany or an SAP affiliate
company.
Intel Xeon, Intel, Itanium, Intel logo, Intel Inside logo, and Intel Centrino logo are
trademarks or registered trademarks of Intel Corporation or its subsidiaries in the
United States and other countries.

Linux is a trademark of Linus Torvalds in the United States, other countries, or both.
SUSE is a registered trademark of SUSE Germany and Novell USA, other countries, or
both. SLES is a trademark of SUSE Germany and Novell USA, other countries, or both.
Red Hat is a registered trademark of Red Hat Limited USA, other countries, or both.
RHEL is a registered trademark of Red Hat Limited USA, other countries, or both.
Other company, product, or service names may be trademarks or service marks of oth-
ers.

Page - 22 © Copyright IBM Corporation 2023

 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

20 Disclaimers

This information was developed for products and services offered in Germany.
IBM may not offer the products, services, or features discussed in this document in
other countries. Consult your local IBM representative for information on the products
and services currently available in your area. Any reference to an IBM product, program,
or service is not intended to state or imply that only that IBM product, program, or ser-
vice may be used. Any functionally equivalent product, program, or service that does
not infringe any IBM intellectual property right may be used instead. However, it is the
user's responsibility to evaluate and verify the operation of any non-IBM product, pro-
gram, or service.
IBM may have patents or pending patent applications covering subject matter de-
scribed in this document. The furnishing of this document does not grant you any li-
cense to these patents. You can send license inquiries, in writing, to: IBM Director of
Licensing, IBM Corporation, North Castle Drive, Armonk, NY 10504-1785 U.S.A.
The following paragraph does not apply to the United Kingdom or any other country
where such provisions are inconsistent with local law: INTERNATIONAL BUSINESS MA-
CHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY
OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS
FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or im-
plied warranties in certain transactions, therefore, this statement may not apply to you.

This information could include technical inaccuracies or typographical errors. Changes

are periodically made to the information herein; these changes will be incorporated in
new editions of the publication. IBM may make improvements and/or changes in the
product(s) and/or the program(s) described in this publication at any time without no-
tice.
Any references in this information to non-IBM websites are provided for convenience
only and do not in any manner serve as an endorsement of those websites. The mate-
rials at those websites are not part of the materials for this IBM product and use of
those websites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes ap-
propriate without incurring any obligation to you.
Page - 23 © Copyright IBM Corporation 2023
 IBM® System Storage™ Architecture and Configuration Guide for SAP® HANA® TDI

Any performance data contained herein was determined in a controlled environment.

Therefore, the results obtained in other operating environments may vary significantly.
Some measurements may have been made on development-level systems and there is
no guarantee that these measurements will be the same on generally available sys-
tems. Furthermore, some measurements may have been estimated through extrapola-
tion. Actual results may vary. Users of this document should verify the applicable data
for their specific environment.
Information concerning non-IBM products was obtained from the suppliers of those
products, their published announcements or other publicly available sources. IBM has
not tested those products and cannot confirm the accuracy of performance, compati-
bility or any other claims related to non-IBM products. Questions on the capabilities of
non-IBM products should be addressed to the suppliers of those products.

This information contains examples of data and reports used in daily business opera-
tions. To illustrate them as completely as possible, the examples include the names of
individuals, companies, brands, and products. All of these names are fictitious and any
similarity to the names and addresses used by an actual business enterprise is entirely
coincidental.

Page - 24 © Copyright IBM Corporation 2023