Wednesday, 10 December 2014

Vendor as a Service offerings

Dear Mr / Mrs Vendor

Please please please - when you come to talk to me about as a service offerings (regardless of IaaS / PaaS / SaaS / TaaS) please have the full story and capability

I am really getting bored of talking technology and not focusing on the service wrapper element ( the clue is in the as a SERVICE phrase)

Everything is possible at the technology layer - service model and operating model is king!

Additionally - when you tell me you have done this for 100's or 1000's of customers before - this should mean that you have a clue... You can help me with operating model definitions etc - please prove and evidence this - as it may allow me to choose you as one of my vendors of choice.

Finally - please do not bullsh!t me - it will become obvious very quickly that you have not done something before / don't understand the service wrapper / can't glue yourself into an operating model (nor help define it) - this alone will at best get you sent to the naughty corner, at worst be asked to never darken my doors again...

Many thanks


Saturday, 6 December 2014

My cloud thought for the day....

Dont try and plug these services into your infrastructure the same way you build / built your legacy systems

- simplify points of integration
- decrease the tooling you consume
- let the vendor use and provide the commoditisted products / tooling / integtration
- make it as simple for yourself

Less is more...


Thursday, 4 December 2014

My thought on cloud services for the day.....

Forget about the technology...

Forget about the funky stuff that happens

get back to the boring stuff....

Operating model... you need to get it sorted to make any of this stuff real within your enterprise..

Just saying.... that is all... 



Monday, 1 December 2014

SSD The Hype Continues

How fast is fast enough??

How power friendly?

What does good enough look like??

Latency times soooooo low.....

These and other statements are things we are use to seeing from the varios makers of these sooper dooper array makers and i get the claims - i really do...

But honestly - what is actually good enough

I get that SSD as a technology is cool, I also get that it can provide a whole bunch of enabling functions as a result of its low-latency characteristics, I get that it has an interesting power profile (sometimes)...

But (and when is there not a but) - Price is still an interesting discussion... There has to be an inflection point of the cost of drives to make this worth considering, OR - array makers, please stop selling at such a big cost point - so that we can marry up value / reward discussions

ok.. and here it comes - the whinge.. Please dont tell me its a lifestyle statement (i am not even sure what this means - but apparently it is!!) and please please please dont tell me that it will make ALL of my infrastructure and applications run far faster - the last time i looked, we hadnt broken the speed of light - so replication for write intensive applications still remains a problem - but NNNOOO apparently not... (Please install wormhole between point a and point b would be in the instruction manual i believe ;-)

So what points am i making....
- Yes i get that SSD tech is good and its here and its gonna rule the world
- There are stll some parts of the tech puzzle for SSD arrays that need to be solved
- Price / Performance / Capability needs to get to a point where it is acceptable in equal measures
- in a lot of cases - the array tech you have on the floor today is good enough - its maybe just a little bulky
- We are probably getting to the last generation or two of spinning disk arrays - but they sure as hell arent dissapearing overnight

does this feel like the "no more tapes" conversation, or maybe "end of mainframe"?