My fondness for Dell M1000e blades continues

Youtube

Chris has been playing with Dell M1000e blade servers this week, and this was the reason for dinner last night. I wanted to hear what he thought.

I have something to confess.

I must be open and get it over and done with.

I’m a fan. I learned about them early last year during some research and was waiting for their release ever since then. Mainly for two reasons, more choice in the market for the end user, and continued innovation of the blade platform, which has to be a good thing.

I don’t have posters on my wall of M1000e blades and haven’t spent evenings learning Dell part numbers.

But I am a fan of the improvements Dell have brought to their new blades and the enclosure.

In particular the ability to backup the enclosure settings to card is very cool. The concept of having to re-configure an enclosure because a part has been swapped need not occur if I can do a simple backup with the card in my laptop/reader.

Chris seems generally impressed with them so far (it’s only been a few days), oh there are a few things he’s been mentioning when comparing the blades, and it tends to be around getting used to these blades as opposed to the IBM/HP ones he plays with.

Anyway they have been looking at Dell’s new Blade FlexAddress for a while now. If you haven’t heard about it, I found this on youtube talking all about it. It’s Dell’s technology which enables you to have a virtual WWN for when you have to replace a blade, so that you can avoid re-zoning the blade to the storage. The other vendors offer similar functionality. Whilst reading up about it, I found this video on youtube, do check it out.

The more we can automate the core every day functions, the lower we can make the support costs and the time to live, the time it takes to respond to and resolve ‘failure’, that by simply switching the blade and pressing power on means we don’t need the storage team to re-zone or re-configure any of the SAN storage, meaning fewer people involved in the resolution, less delays.

Could the work be carried out by operators following specific instructions with the Windows/Unix guys getting called to check it afterwards?