Vendor lock-in, severity across cloud architectures

Several people have asked me in the recent past about the vendor-lock-in issue with various deployment architectures. The below picture should be self-explanatory on the amount of lock-in across SaaS, PaaS and IaaS. The lock-in that I am referring-to here is a combined aggregate of Vendor lock-in + data lock-in + dev environment lock-in + data lock-in.

Best practices for zoning in SAN

Ever wondered if the same best practices of SAN zoning are applicable with or without server virtualization? Let me give some pointers to create a stable, manageable & secure SAN zone.

• Always go for Zoning, even if LUN Masking is being used
• Always implement a default zone
  
• Disable any unused storage ports on the switch to increase security and avoid potential problems
• Use pWWN identification for all Zoning configuration unless D,P identification is required
  
• Create Zoning aliases and names with only as long as required to allow maximum scaling
• Single Initiator Zoning are to be used with separate zones if a HBA is carrying both types of traffic (eg for tape and disk traffic)
• Use accurate Zoning terminology
  
• Describe Zoning by enforcement method and identification type
• Always use the vendor given software to validate the zoning configuration
• Zones should use frame-based hardware enforcement