The other day I was talking to Khalid about Mr. Crofford’s blog regarding power consumption in our datacenter and he mentioned that we may cut down on cost (power/datacenter space) if we get rid of the redundancy. Later that day while talking to Rusty and Adrian, we came to conclusion that in some of our accounts, how active-active network configuration for load sharing/load balancing has added more complexity.

These made me think – which I rarely do. J I did a quick MTTR (mean time to repair) and MTBF (mean time between failures) calculations for Cisco devices and it seems that we can easily meet an

Now imagine a specific cloud architecture (Infrastructure 3.0 or now being called IAAS – infrastructure as a service) which has less redundant elements which will result in a huge power savings over the year.

We can come up with processes and very tight configuration management (as per ITIL v3) that will allow us to swap out a failed piece of hardware quickly and meeting our SLAs.

Moreover if we have the right procedures, every two months or quarter we can bring up the cold standby and seamlessly failover from the active component thus making sure that the configurations are sync’d and not wearing out the same piece of equipment with overuse.

In this area the ongoing support cost (read human resource) will go up. But this may spur us to do more automation.

If anybody is interested and time permitting we can indulge in a full study. This may be one of the area where we can save some money for us and our clients by going green. We will do good both to our customer and to our future generations – what a thought!!

******Time to create few more Spore creatures to inhabit my Spore Ecoverse.********