[SAP HANA Academy] SDA: SAP HANA Spark Controller Overview [SPS 10]

ESXi Host

UCS Service Profile

The ESXi host provides the platform where virtual machines are running on. The service profile contains the configuration of the hardware in a Cisco UCS environment. Service Profile Templates or vSphere Auto Deploy can be used to ease the ESXi deployment process. In this example, a standalone service profile creation is shown.

For each vSwitch, it is recommended to configure two uplink interfaces with MTU 9000 as trunk. The VLAN assignment takes place in the port group configuration of the vSwitch.

In order to get the best performance for virtualization, certain BIOS features should be enabled. The c-states can be controlled by the hypervisor and do not necessarily have to be disabled. It depends on the performance needs vs. power saving aspects how balanced this configuration should be.

VMware vSphere screenshot

Although the use of VM-FEX is optional, it is recommended to enable all Intel Direct IO features.

Network

SAP HANA has different types of network communication channels to support the different SAP HANA scenarios and setups. It is recommended to consult the SAP HANA TDI - Network Requirements whitepaper.

Source: SAP AG

On the basis of the listed network requirements, every server must be equipped with two 1 or 10 Gigabit Ethernet (10 Gigabit Ethernet is recommended) interfaces for scale-up systems to establish communication with the application servers and users (client zone). If the storage for SAP HANA is external and accessed through the network, two additional 10 Gigabit Ethernet or 8-Gbps Fibre Channel interfaces are required (storage zone). Scale-out systems need a 10 Gigabit Ethernet link for internode communication (internal zone). When using multiple ESXi hosts in a vSphere Cluster with enabled DRS, at least one additional 10 Gigabit Ethernet link is required for vMotion traffic.

For the internal zone, storage zone and the vMotion network, it is recommended to configure jumbo frames end-to-end.

The network traffic can be consoliated by using the same vSwitch with several load-balanced uplinks.

Storage

The storage system must be certified as part of an appliance, independent of the appliance vendor, or must be certified as SAP HANA TDI storage.

It is recommended to physically separate the the origin (VMFS LUN or NFS export) of the datastores providing data and log for performance reasons. Following performance classes can be distinguished:

Category	Read Performance	Write Performance
OS boot disk	medium	low
/hana/shared	medium	low
/hana/data	very high	high
/hana/log	high	very high
backup	low	medium

It is also recommended to consult the recommendations from the storage hardware partners:

EMC: Tailored Datacenter Integration Content

NetApp Reference Architecture: SAP HANA on VMware vSphere and NetApp FAS Systems

NetApp Configuration Guide: SAP HANA on NetApp FAS Systems

______________________________

VMware: Best Practices and Recommendations for Scale-Up Deployments of SAP HANA on VMware vSphere

In the next part of theSAP HANA Academy’sLive4 ERP Agile Solutions in SAP HANA Cloud Platformcourse,Tahir Hussain Babar (Bob)provides an overview on the source Hadoop system used in the course. The Hadoop system is where the EPA (Environmental Protection Agency) weather data has been loaded. It will be a source data set for a SAP Fiori application in the SAP HANA Cloud Platform. The data doesn't need to be replicated in HCP and thus the data is accessed using virtual tables and remote sources created with SAP Smart Data Integration. Watch Bob’s video below.

(0:30 – 2:30) Background on Data in Hadoop

In the Live4 course a connection will be made from HCP to a Hadoop data lake that contains over 70 years of weather data from the EPA. There is no need to copy all of that weather data into HCP so a virtual table for that data lake will be created so we will only get the data that is needed.

Many options exist for data connectivity. One way demonstrated earlier in the course was to use HCI-DS is to copy the data to the remote system. Another method is to use SAP Smart Data Integration to replicate the data. With SDI, once the data is changed in the source system it will immediately be reflected in HCP. Yet another method is using virtual tables. With virtual tables no replication is necessary as you have a virtual view in your data source.

(2:30 - 7:00) Examination of the Databases in the Hadoop system

On a Windows machine Bob is running Hadoop Hortonworks data platform 1.2. If you don’t have your own Hadoop system please check out this series of four videos on the SAP HANA Academy on how to use, install and configure Hadoop Hortonworks.

This data set (download here on GitHub) contains weather data from the EPA. Everyday the EPA surveys over 300 weather stations and has recorded various measurements for over 70 years. Instead of storing all of that data on the hot storage system of SAP HANA we will store it on Hadoop as a cold storage system that we can then access.

Bob launches a command prompt window and connects to his HIVE 0.12.0 system and then executes hive from his bin directory. If you're following along with the course please make sure to use HIVE 0.12.0 throughout. Bob has already loaded the EPA databases into his Hadoop system so by entering show databases; as a command he can view his three databases (default, epa and live2). Then entering show tables in epa; shows the five tables from the EPA Bob has exposed in this database (aqi_datalake, humidity_datalake, pressure_datalake, temp_datalake and wind_datalake).

Next Bob enters select * from epa.aqi_datalake; to see all of the 70 years worth of air quality index data stored in a table format that has been translated using the HIVE query language. Hive query language enables BI tools and SAP HANA to understand the format.

In the next video Bob will show you the user authorizations and agents you need in order to connect your Hadoop system to HCP using Smart Data Integration.

For further tutorial videos about the ERP Agility with HCP course please view this playlist.

SAP HANA Academy - Over 1,200 free tutorial videos on SAP HANA, Analytics and the SAP HANA Cloud Platform.

Hello All,

Its been some time that I have been working in HANA and related areas like SLT, Lumira, Fiori and so on.

So thought of sharing some topics here, which would be handy.

Disclaimer :

1) This series is exclusively for Beginners in HANA and all those HANA experts here, please excuse me

2) These are some Solutions/Observations that we have found handy in our projects and am quite sure there would be multiple ways to derive the same result.

3) These series of documents are collaborative in nature. So please feel free to edit the documents,wherever required!

4) All the points mentioned here is observed on HANA systems whose revision is >= 82.

Part 1 of this series can be found here --> Tips, Experience and Lessons Learned from multiple HANA projects(TELL @ HANA - PART 1)

Part 2 of this series can be found here --> Tips, Experience and Lessons Learned from multiple HANA projects(TELL @ HANA - PART 2)

Part 3 of this series can be found here --> Tips, Experience and Lessons Learned from multiple HANA projects(TELL @ HANA - PART 3)

34)Related to HANA

Use Case: We were unable to access/open the Catalog Folder of our HANA instance and the following error message was appearing.

Solution: We had raised the issue with SAP Internal Support.

And as of their observation: all HDB processes were online and there wasn't any obvious reason for an error.As a Quick fix, they restarted the HANA DB and

the error was solved.(our's was a Demo System anyway)

Note: Some related Discussions --> https://scn.sap.com/thread/3729403

35) Related to HANA Studio:

Use Case: We had to uninstall the Existing version of HANA Studio.

Solution:

Got to Control Panel --> Uninstall the HANA Studio.

The Lifecycle Manager will ask you to enter the HANA Studio installation instance(in our case; it was 0)

After entering 0, you will get the following screen:

By pressing any key, you will get the message that the HANA studio version is successfully uninstalled.

36) Related to HANA

Use Case: At times, while navigating through the HANA Contents, we come across the across the following message(Contains Hidden Objects).

Solution:

Go to the Preferences --> HANA --> Modeler --> Content Presentation --> Check 'Show all Objects'.

Now the hidden objects will be displayed:

37) Related to HANA SQL

Some Useful SQL Commands:

a) Renaming a column of an already existing Table:

RENAME COLUMN "<Schema_Name>"."<Table_Name>"."<Old_Column_Name>" to "<New_Column_Name>"

b) Adding a new column to an already existing Table:

ALTER TABLE "<Schema_Name>"."<Table_Name>" ADD (<New_Column_Name><DataType>)

c) Update a Column Entry:

UPDATE "<Schema_Name>"."<Table_Name>" SET "<Column_Name>" = '<New_Entry>' where "<Column_Name>" = '<Old_Entry>'

d) IF Function:

If("Status"= 'Completed Successfully','0',if("Status"= 'In progress','1',if("Status"= 'Withdrawn','2','?')))

38) Related to HANA Studio:

Use Case: We got the following error while previewing an HANA Analytical view:

Message: [6941] Error Executing Physical Plan : AttributeEngine: this operation is not implemented for this attribute.

Solution: The above error message point towards a field named CMPLID.

On careful observation, it was found that the CMPLID has different data types in the connected attribute views and the Fact table in the following analytic view.

Related SAP Note: 1966734 - Workaround for "AttributeEngine: this operation is not implemented for this attribute type" error

39) Related to HANA:

Use Case: How to find the Schema Owners?

Solution: SELECT * FROM "SYS"."SCHEMAS"

40) Related to HANA:

Use Case: How to provide specific authorizations to some limited tables within a schema?

Solution: Object Privileges --> Schema Name --> Right Click --> Add catalog objects --> Provide the specific table names.

Hope this document would be handy!

41) Related to sFIN

Use Case: The definition of HANA Calculation View BANKSTATEMENTMONITOR is not correct after migration to SAP Simple Finance, on-premise editon 1503.

The expected definition of HANA view after migration is something like the following:

Unfortunately due to a program error, the view definition after migration will still look something like the following:

Solution: We had raised this issue with the support/development team and they have now released the following new OSS note.

2205205 - Bank Statement Monitor: Definition Correction of HANA Calculation View BANKSTATEMENTMONITOR

After following the manual activities mentioned in the note, the issue will be resolved.

42) Related to SLT:

An SLT configuration was already created without giving multiple usage option.(You want to switch from 1: 1 to 1: N in already existing SLT configuration)

No we wanted to create a new connection with the same source and a different target, but system was not allowing us to do the same, as we were getting the message that a configuration with the same source is already available.

Solution: NOTE 1898479 - SLT replication: Redefinition of existing DB triggers.

The solution for this issue was explained in the note and the manual steps (1-9) has to be done in the SLT system to solve this.

Will keep adding more points here...

Prabhith-

Virtual Machine

CPU and Memory Sizing

This document is not a sizing instrument but delivers a guideline of technically possible configurations. For proper sizing, refer to BW on HANA (BWoH) and Suite on HANA (SoH) sizing documentation. The ratio between CPU and memory can be defined as 6.4 GB per vCPU for Westmere-EX, 8.53 GB per Ivy Bridge-EX and 10.66 GB per Haswell-EX vCPU. The numbers for Suite on HANA can be doubled.

See that the ESXi host and every virtual machine produce some memory overhead. For example, an ESXi server with 512 GB physical RAM can not host a virtual machine with 512 GB vRAM because the server would need some static memory space for its kernel and some dynamic memory space for each virtual machine. A virtual machine with eg. 500 GB vRAM would most likely fit into a 512 GB ESXi host.

This table shows the vCPU to vRAM ratio on a Haswell-EX server. Under "VMs per 3 TB Host", the maximal amount of VMs is shown considering two scenarios:

BWoH prod
- This is the amount of virtual machines that can be deployed on one host for BW on HANA in a productive environment, that means, there is no resource overcommitment.
CPU overcommit
- This is the amount of virtual machines that can be deployed on one host while allowing CPU overcommitment. Memory overcommitment is not recommended at all for HANA virtual machines.

vCPU		vRAM	VMs per 3 TB Host
BWoH	SoH	vRAM	BWoH prod	CPU overcommit
18	18	64 GB	8	46
18	18	128 GB	8	23
36	18	256 GB	4	11
36	18	384 GB	4	7
54	36	512 GB	2	5
72*	36	768 GB	1	3
108*	54	1 TB	1	2
-	72*	1.5 TB	-	1
-	108*	2 TB	-	1

* 4 TB vRAM and 128 vCPU per VM with vSphere 6

More details are available here:

SAP: SAP HANA Guidelines for running virtualized

VMware: Best Practices and Recommendations for Scale-Out Deployments of SAP HANA on VMware vSphere

Virtual Machine Creation

Create virtual machine with hardware version 10 (ESXi 5.5)

Select SLES 11 64-bit or RHEL 6 64-bit as Guest OS

Configure cores per socket

It is recommended to configure the cores per socket according to the actual hardware configuration, which means:

10 cores per socket on HANA certified Westmere-EX processors
15 cores per socket on HANA certified Ivy Bridge-EX processors
18 cores per socket on HANA certified Haswell-EX processors

Virtual SCSI controller configuration

Configure 4 virtual SCSI controllers of the type "VMware Paravirtual".

Configure virtual disks

To fully utilize the virtual resources, a disk distribution is recommended where the disks are connected to different virtual SCSI controllers. This improves parallel IO processing inside the guest OS.

The size of the disks can initially be chosen lower than the known requirements for HANA appliances. This is based on the capability of in increasing virtual disk size online.

Disk	Mount point	Size (old measure)	Size (new combined measure)
root	/	80 GB	80 GB
hanashared	/hana/shared	1 * vRAM	1 * vRAM
hanadata	/hana/data	3 * vRAM	1 * vRAM
hanalog	/hana/log	1 * vRAM	vRAM < 512 GB: 0.5 * vRAM vRAM ≥ 512 GB: min. 512 GB

Configure virtual network adapters

Configure one VMXNET3 adapter for each network and connect it to the corresponding port group. See that some networks are only configured on ESXi level, such as storage and backup network.

Enable Latency Sensitivity settings

To ensure the latency sensitivity, CPU and memory reservation has to be set, too. While the CPU reservation can vary, the memory reservation has to be set to 100 %. Check the box "Reserve all guest memory (All locked)".

After creating a VM, the guest OS installation can begin.

______________________________

Guest Operating System

Installation

The installation of the Linux OS has to be done according to the SAP Notes for HANA systems on SLES or RHEL. The network and storage configuration parts are heavily depending on the TDI landscape, therefore no general advice can be given.

Disk Distribution

The VMDKs should be distributed on differently tiered storage systems because of performance reasons. Example:

In the end, a realistic storage quality classification as well as a thorough distribution of the disks among datastores and virtual SCSI adapters ensures good disk IO performance for the HANA instance.

Configuration

With regards to the network configuration, it is not recommended to configure bond devices inside the Linux guest OS. Such a configuration is used in native environments to guarantee availability of the network adapters. In virtual environments, the redundant uplinks of the vSwitch take on that role.

In /etc/sysctl.conf some tuning might be necessary for scale-out and in-guest NFS scenarios:

net.ipv4.tcp_slow_start_after_idle = 0

net.core.rmem_max = 16777216

net.core.wmem_max = 16777216

net.core.rmem_default = 262144

net.core.wmem_default = 262144

net.core.optmem_max = 16777216

net.core.netdev_max_backlog = 300000

net.ipv4.tcp_rmem = 65536 262144 16777216

net.ipv4.tcp_wmem = 65536 262144 16777216

net.ipv4.tcp_no_metrics_save = 1

net.ipv4.tcp_moderate_rcvbuf = 1

net.ipv4.tcp_window_scaling = 1

net.ipv4.tcp_timestamps = 1

net.ipv4.tcp_sack = 1

sunrpc.tcp_slot_table_entries = 128

MANDATORY: Apply generic SAP on vSphere optimizations and additional HANA optimizations for SLES and RHEL.

Validation

To validate the solution, the same hardware configuration check tool as for the appliances is used but with slightly different KPIs. SAP supports performance-related SAP HANA issues only if the installed solution has passed the validation test successfully.

Volume	Block Sizes	Test File Size	KPIs
Volume	Block Sizes	Test File Size	Initial Write (MB/s)	Overwrite (MB/s)	Read (MB/s)	Latency (µs)
Log	4K	5G	-	30	-	1000
	16K	16G	-	120	-	1000
	1M	16G	-	250	250	-
Data	4K	5G	-	-	-	-
	16K	16G	40	100	-	-
	64K	16G	100	150	250	-
	1M	16G	150	200	300	-
	16M	16G	200	250	400	-
	64M	16G	200	250	400	-

Source: SAP AG, Version 1.7

______________________________