Explore two-step migration - lift and shift followed by DMO

Completed

Consider the following for the two-step migration:

  • Ensure connectivity to Azure is available via Express Route (recommended) or VPN.
  • Allocate the target infrastructure on Azure, which includes the cloned system and target SAP NetWeaver and SAP HANA database servers. The Azure infrastructure can be deployed using predefined Azure Resource Manager templates.
  • The cloned system can be built with a homogeneous system copy (backup/restore) or via DBMS replication tools (e.g. Oracle Data Guard or SQL Always-On).
  • Business and technical testing should be initiated (with functional, integration and acceptance testing to ensure the move of data has been successful).
  • Following business and technical validation, the traditional Database Migration Option (DMO) process can be followed to migrate and upgrade to SAP HANA.
  • DMO can be used with the memory pipe method (for example, export/import occurs within the same application server and memory segment for accelerated migrations).
  • Following the migration to SAP HANA, business and technical validation should again be initiated.
  • In this approach, two downtimes and two testing cycles are required.

DMO optimization options

There are many factors that influence the downtime associated with an SAP database migration and include (amongst others):

  • Scope: Software changes, Unicode conversion, data center relocation
  • Source system performance: CPU, I/O, memory, DBMS performance, SAP NetWeaver release
  • Source system database size: Database size, largest tables, status of housekeeping
  • Target system performance: CPU, I/O, memory, DBMS performance, SAP NetWeaver release
  • Network: Network speed, bandwidth, latency
  • Toolset: SAP Software Provisioning Manager (SWPM), Software Update Manager (SUM), DMO of SUM, version of the toolset
  • Migration approach: Standard vs Minimize Downtime
  • Ramp-up/ramp-down activities: Interface management, batch job management
  • Pre and post migration validation: Functional, integration, and acceptance testing.

Broadly speaking, the potential for performance optimizations can be seen in the following three areas:

  • Export
  • File Transfer
  • Import

Infrastructure/hardware

The infrastructure/hardware reduction lever consists of the following:

On-premises

  • Deploy a dedicated migration server (PAS / AAS) with extensive compute capability to execute SUM
  • Deploy Solid State Disks (SSD) on the source database and migration servers.
  • (PAS / AAS)

Microsoft Azure

  • Express Route connectivity is recommended with the maximum bandwidth available (currently available up to 10 Gbps)
  • use the dmotocloud.sh script (RSYNC) to move files from an on-premises source to an Azure target for the DMO System Move Option with Parallel Transfer.
  • Implement storage snapshots for time-efficient backups during the cutover period.

Key considerations for downtime optimization for DMO

  • Source database optimization (index rebuild, DB statistics, DB parameters, DB file system). A useful reference can be found in the table below.
Database Reference
Oracle SAP Note #936441
DB2 DB2 Optimization Techniques for SAP Database Migration And Unicode Conversion
SQL Server SAP OS/DB Migration to SQL Server–FAQ v6.2 April 2017
  • Use the latest version of migration tools, R3*, kernel, etc.
  • Network parameterization (e.g. MTU settings, reset counters)
  • OS-related parameterization (e.g. Q depth)
  • SAP HANA database parameters (related to save points, log mode, and timeouts)
  • DMO Migration parameters (e.g/ duration files, manual table splits, manual table sequencing order, fast load for BLOB, etc.)

DMO Benchmarking

  • Used to capture export and import phases for a subset of data. It can be configured to run on specific tables.

Repeat multiple runs of downtime phase

  • Running the execution phase multiple times is possible by reworking the table splitting activity (thus optimizing table splitting that will reduce the downtime of the DMO procedure).