Stuiesav blogs about EMC’s positioning of VMAX vs VNX and the problems that EMC marketing appear to have with putting clear blue water between the products. This has been a long running problem which started with the purchase of DG and the introduction of the Clariion range into the EMC portfolio leading to some serious internal conflict within EMC between competing product teams.
It did appear however that peace had more or less broken out in the camps but I wonder? As VNX continues to grow in capability and scalability, the high-end product operates a more and more rarified atmosphere and there is less requirement for it. And like other companies before, EMC are at risk at some quite serious internal compete; like IBM when often you would find AS/400, RS/6000 and mainframe all pitched into a user to do the same job, there is clear and pleasant danger of this happening here.
Actually EMC have a tremendous amount of overlap across most of their product-range at the moment, VMAX overlaps with VNX which overlaps with Isilon which overlaps with Atmos which overlaps with Centera; there’s not a huge amount of external coherence.
The senior-guys, the marketing gods and the bloggers within EMC will pitch this as strength in diversity but will the sales-guys get it and will they be able to successfully pitch or will they simply concentrate on the products which generate the most revenue and get them closer to their target? If I’m being cynical, I’d expect the account manager to try to get VMAX through the door in preference to VNX but if there was a risk of a customer going NTAP/HP/IBM, then I’d pitch in with VNX.
And of course, I’m aware that I’ve not mentioned one of EMC’s platforms; VPLEX. I suspect like many, I’m wondering what is going on with VPLEX and what direction it is going in. It’d be nice to see The Storage Anarchist put together a blog on VPLEX and VPLEX directions? But perhaps it’s simply easier to trash-talk the opposition and blog about meaningless benchmarks?
(Marc Farley from HP here) I too am suddenly aware of the issues regarding overlapping products – something we didn’t worry about at 3PAR. To answer your question, sales people sell what they can to earn the most and that depends on the commission schedule and the relative ease that something can be sold. Companies invest a lot of money and time trying to satisfy multiple product groups and keep them all competitive. Product positioning is not always clear – or necessarily logical where overlap occurs. That’s just a fact of life in this consolidating industry. FWIW Netapp will probably have to start dealing with this issue too, now that they are acquiring Engenio.
As for Vplex, it’s not really a platform, which is why it doesn’t matter all that much to EMC or their customers. It’s not how anybody butters their bread.
Since I was with a customer who virtualises all their entire storage storage estate with VPLEX a few days ago and is using it to teleport their VMs around the place they’d disagree with Marc.
VPLEX was introduced at EMC World 9 months ago. You’ll be hearing about it again soon enough.
Marc knows of which he speaks – HP reps are reportedly being paid 2x commission for selling 3PAR kit instead of Hitachi kit, even after the price hike Donatelli injected on 3PAR’s lineup.
The road to 3PAR success at HP is lined by sales comp plans (pity poor Hitachi – first they lose Sun, and how HP’s giving them the cold shoulder).
Oddly, we don’t see any of this so-called “confusion” when auto manufacturers have overlap…none of this “which is better” between Toyota/Lexus, Nissan/Infinity, Honda/Acura, Chevy/Olds/Cadiallac/Buick, Ford/Mercury, VW/Audi/Porsche, etc. etc. etc. – even when they are REALLY the same cars in both lineups, just one has nicer leather!
VNX/VMAX differentiation lies in things like multi-controller, availability, performance, SRDF, connectivity, management, automation, scale, efficiency, capacity, workload optimization and more. There is no simple algorithm to decide, I’ll agree, but that hasn’t stopped practically ANY supplier in ANY industry from offering overlapping choice.
Mach. Ado. Over. Nothing.
IMHO!
I was wondering when somebody was going to pitch the tired old auto manufacturer analogy.
I just want to be clear in my understanding as to what Barry is saying; are you saying that VNX and VMAX are the same but one has nicer leather than the other?
I actually find it interesting when this one gets trotted out again and again. No-one complains at confusion in the automobile industry because when it boils down, what car you drive says more about how much money you want to waste and what message you want to give out about yourself. If you are buying storage to try to reflect how important you think you are and how much money you want to waste…then we have hell of a problem.
I reject the notion that having more than one product confuses anything. If you can buy washing detergent from the shelves stacked with different powders, liquids, gels and sprays you can buy storage.
If a customer wants Unified Storage (Block & File) it’s VNX. If they require Scale Out Block or Scale Out NAS it’s VMAX or Isilon. If they require compliance for long term retention of data it’s Centera and if they require cheap and dense policy based object storage it’s Atmos.
This isn’t some great academic exercise. The primary question is always, ‘What are you looking to do?’
To anyone that actually compares the VNX/CX4 line of EMC arrays and the VMAX it’s very clear and obvious what the differences are. There is very little overlap in those product lines. An organization’s requirements may overlap the lines, but the lines themselves do not. Often I have the conversation with a customer about whether they really fit in to a VNX or in to a VMAX. What are your requirements for resiliency and scalability as well as budget. That usually bears it out pretty quickly.
Outside of Ficon connectivity which is the one capability which forces one into making a decision to go VMAX; I do really wonder how often that VMAX is sold when a VNX is actually a good-enough answer?
Your job is not to spend all of customer’s budget…sometimes it feels like vendors believe that they have a duty to get as close to a customer’s maximum spend as possible…
Yes, when provoked there is always a response from EMC, so now we are all anxiously awaiting the blitz of Vplex successes, such as the ability to cram more of them into a Mini than Filers. It is and always will be a bump in the wire similar to SVC.
Buying storage is like buying soap? Only if you have a stink that is so bad, you need to do the quickest thing possible to cover it up.
It is true that looking from the outside (as I was before) there’s this stupendous number of products offered by EMC. Having said that, it’s not the customers job to know all the products (that’s my job as a Technical Consultant/architect/etc.) but rather know their business needs/requirements.
Given that, I can usually focus down on what the right product or product mix is (although we do NOT start with products…always better/best to focus on the business requirements/processes/etc. and refine those as much as possible during discussion. Yes, there is overlap (inevitable result of acquisition) but at the same time each product has one or several specific strengths/use cases. What I like is that I do often have multiple options within the same product segment (ex. Backup…Data Domain, Avamar, Networker, etc.) so can get much more fine grained in my recommendations (i.e. not stuck with the “everything looks like a nail since I only have a hammer” syndrome).
At the same time, there is gradual consolidation on the backend of the various products…keeping the various strengths (hopefully) but combining at various levels. One example would be what’s been done with the VNXe line under the covers. Another is cross-pollination between Avamar and RecoverPoint.
vPlex is all about storage virtualization — think IBM SVC but with much bigger dreams (i.e. vMotion between datacenters via datastores that are active/active — Metro distances today, Geo distances tomorrow….live demos done at EMC World actually).
And….all of this stuff existed before acquisition anyway (and had to be doing somewhat well to be acquired) so I’d say the confusion level is roughly about the same actually. 😛
I would like to welcome both Andrew and Jason to my blog; both work for an EMC partner, just so we are all clear to where people are coming from 🙂
Ah…thanks for that. I implied that (can find it off my blog/public LinkedIn too) but you’re right I didn’t directly reference my employer association…thanks. Not trying to keep any secrets…just post my current thoughts. 🙂
[Disclosure, I work for EMC]
I’ve heard this drum beat since the DG acquisition, but its not usually from customers. Is there overlap between VNX and Symm? Sure. The deciding factor for must customers to move to symm is either massive scale or 100% uptime requirements that any dual controller architecture just can’t provide. Or FICON/ESCON as you pointed out.
I think with the Engenio acquisition last week, we can finally put the nail in the ‘one size fits all’ coffin.
Sym provides 100% uptime? Really? That would be one heck of an announcement.
And you might not hear it from customers but are you listening? It does get talked about and there are questions about when to deploy VNX and when to deploy VMAX. It might be an interesting post to discuss various scenarios and actually come up with some answers? Some worked examples where you think one or the other makes sense.
interesting the amount of debate that this blog entry has caused. I whole heartedly agree with Martin’s summary (but thats not a suprise given that i posted a similar blog).
I do believe the product lines are blurred, and that the confusion really sets in when a customer that doesnt live in storage world needs to make a decision. The article i wrote is based exactly on that scenario, and the fact that a common vendor was pushing different answers just shows that confusion does not just exist on the customer side but also within the vendor that is selling (anything they can?).
Use requirements to select a product, dont just push every product so that the customer buys one of your offerings… (and i see more than one vendor use this approach….)
(for those interested – my entry sits here: http://www.stuiesav.com/2011/03/blurred-marketing-emc-product-line.html
cheers,
S.