SVC to V7000 Migration

Hello all, for those of you wondering, yes I am still alive! I apologise for going AWOL for 6 months! However I am now back and hope to be regularly updating my blog with all my experiences and findings e.t.c. 

So whats the story? Back in 2010 IBM released the Storwize V7000 and subsequently the Storwize V7000 Unified. The IBM Storwize V7000 is IBMs next generation mid-range storage Virtualization platform. The platform has been built utilising many features from IBMs best of breed solutions including SVC, DS8000 and XIV. The features list is huge and my favourites include Easy Tier, the Dynamic Migration capabilities, the External Storage Subsystem Virtualization capabilities (these are not vendor limited - V6.3 Supported Hardware List, Device Driver, Firmware and Recommended Software Levels for IBM Storwize V7000) VAAI integration, vCenter integration and Clustered Systems. In a sentence it's a solid mid-range storage subsystem containing many features you would only expect to see on an enterprise level storage subsystem.

Data migration from an SVC presenting a DS3k, DS4k & DS5k to a V7000. It looks simple doesn't it? However I recently came across a scenario with an environment running old 4F2 SVC nodes and planned to migrate all data under the SVC on DS3k, DS4k and DS5k storage onto a V7000. This presented a problem in that the V7000 can only be presented as a host to an SVC running 5.1.0.8 and above - V5.1.x - Supported Hardware List, Device Driver and Firmware Levels for IBM SVC A quick check over the Node and UPS Interoperability Matrix for SVC shows that the 4F2s are only compatible with SVC code level 4.3.x. 

VMFS-3 >> VMFS-5

The first new feature I am going to focus on in vSphere 5 is VMFS-5. As you know within VI3.x & vSphere 4.x VMFS-3 was used. So whats changed?  VMFS-5 offers the following enhancements:

• A unified 1MB file block size
• 64TB datastores without increasing extents. This is a massive increase on the previous 2TB single extent datastore limit using VMFS-3
• Smaller Sub Block now uses 8KB instead of 64KB. Reducing the amount of disk space being consumed by small files
• VMFS-5 uses the file descriptor location in the metadata for storage rather than files blocks. When files grow above 1KB these files will then start to use the new 8KB sub blocks, resulting again in reduced amount of disk being consumed by very small files.
• Support for more than 100,00 files in a datastore, that's three times the number of files supported within a VMFS-3 datastore  
• vStorage API for Array Integration (VAAI) hardware acceleration locking used more extensively improving file locking performance over VMFS-3

Wow, so how do you migrate from VMFS-3 to VMFS-5? Those that performed VMFS-2 to VMFS-3 upgrades will remember remember the complex scripts that required running to perform the upgrade. However I am happy to confirm the upgrade from VMFS-3 to VMFS-3 is non disruptive, easy to perform, takes only a matter of minutes and can be simply completed via the vSphere 5 Client (see screenshots below)