back to article Cisco and NetApp fatten up Flexpod, chuck it at cloud biz

Cisco and NetApp are drawing closer together with their FlexPod converged system effort, an attempt to grab branch office, data centre and public cloud infrastructure business. Cisco has servers and networking, which is not enough, and NetApp has storage, which is not enough either - because IT people need servers, storage and …

COMMENTS

This topic is closed for new posts.
Pint

Why not just buy a fully converged and integrated stack from one Supplier?

Huawei Enteprise systems already offer a fully converged Cloud stack with its IT product range and specifically its fusionCube products.

Servers, Storage, Switches, Racks, UPS, Hypervisor, Cloud Management, Cloud Portal. All integrated and from a single supplier :-)

They can scale to equivalent vBlock and Flexpod configurations, and are carrier class having been implemented in many Asia Telcos..

best regards ;-) Malcolm

0
2
Bronze badge
Thumb Down

Re: Why not just buy a fully converged and integrated stack from one Supplier?

The idea of buying converged systems from a single supplier is often pooh-poohed as "proprietary", especially by suppliers who don't have the three technologies needed in-house, and the main three suppliers in that position are Cisco, EMC and NetApp. Both EMC and NetApp are trying to attract the attention of Cisco, the great converged stack prize, and hoping to be chosen as its preferred partner.

I think that might answer your question.

The problem with the "converged stack" theory is that it's the mainframe redux: you're locked into buying giant units of equipment from a single vendor. Virtualization ameliorates this issue somewhat, insofar as you can easily move your processing workloads elsewhere, but storage lock-in is especially pernicious since storage is the hardest resource to move away from. The discerning IT equipment purchaser will look for the opportunity to retain flexibility.

Also, take your Huawei shilling elsewhere.

1
0
Anonymous Coward

Re: Why not just buy a fully converged and integrated stack from one Supplier?

I agree that what you're creating is a mainframe like lump of 'compute' but you're doing it using small, modular, open components rather than expensive proprietary mainframes. 'Converged stacks' from vendors like HP allow you to integrate non-HP elements into the overall architecture so deliver far from the lock in scenario you try and portray. And just as you can move processing workloads around easily these days, so you can with storage behind virtualised environments with things like storage vMotion. Traditional workloads create a much bigger migration headache.

0
0
This topic is closed for new posts.

Forums