Insider’s Perspective: Ionix and VMware

The sub-title of my blog is “an insider's perspective, technical tips n' tricks in the era of the VMware Revolution”.  I’ve said it before, I’ll say it again – I think I’ve got the coolest job on the planet :-)   I’ll give you a couple “insider” hints.   2010 will be a very big year on the VMware front – loads of big releases, new functionality and expanding use cases.   Likewise, it’s shaping up to be a block-buster year on the product front at EMC.   The number of “headline” level things down the storage and security fronts between now and EMC World (May) are huge.   Between now and the end of the year – massive.

So today, a big move occurred on the management side of the house that’s been underway for a while.   You can read more here. For more detail, and that “insider perspective”, read on…

My blog posts are long, and I’m long-winded by nature – so here’s the conclusion before the detail:

This is a move where VMware in an instant, makes a quantum increases the depth of resources, tools and capabilities in things that are focused on server-layer and up.   This expands the vCenter family of management capabilities, and the teams that develop them, and the teams take them to market.

This is a move where EMC, in an instant, can focus ongoing management M&A and R&D and go to market on the things that focus on UCS, networks, and storage - the server layer and down.  This accelerates EMC’s management efforts.

Ok – now for more detail…

Before going further, understand: EMC and VMware do indeed operate as “arms length” technology partners. 

The model has VMware operating independently, and EMC partnering in a way that is analagous to our partnership model with Microsoft.   We know that this dynamic is very important for VMware’s continued success.

What is perhaps different versus the partnership models with others is the degree of shared vision – and that we have linked success.

VMware’s strategic goal is the journey to the private cloud.   EMC’s strategic goal is the journey to the private cloud.   One of Cisco’s strategic goals is the journey to the private cloud.

I know that many look at that statement - “journey to the private cloud” and write it off as marketing.   Well it is, in a sense.   It’s also a massive rallying call to our resources – to be focused on a strategic goal.   The other thing to understand is that it’s not some future state, many elements are possible today.

Taking it down a nerd notch, what exactly do we mean?

Enabling cloud operating models (both technologically and economically) inside datacenters and service providers.  Doing that for all use cases (servers AND clients,  legacy apps AND next-gen “built for cloud” apps).    All of which is possible today (look at Terremark, or look at your own virtualized datacenter) – and every thing we do going forward is about expanding this set of use cases, making it easier, making it more “out of the box”, and more cost-effective and scalable.   Ultimately where we are trying to go are models where the internal clouds and the external clouds can transparently blend together into hybrids, or Private Cloud models.

This model – technology partners with shared vision, shared goals - means that the VMware/Ionix news today isn’t an “internal restructuring”, but in fact a strategic decision by two companies to align their mutual assets for focus and execution.  

• EMC came to the conclusion that since VMware is more and more the context for our customers, and people look to VMware for the server layer and up, and EMC to the infrastructure and down – some of the assets would get more leverage outside EMC.
• VMware came to the conclusion that they can accelerate their vCenter management stack capabilities, and selected the EMC assets that they felt could accelerate them the most.

Want the next level of detail?  Keep reading…

EMC has working to assemble a “best of breed” – but specifically targeted - management portfolio, outside “traditional” Storage Management that people naturally look to us for.  Why?  Because - we think management in this target “cloud” state that our customers are looking for, well “management” disciplines are going to change, and the technology requirements will too.   I’ve said it before, and will say it again.  We’re not trying to beat HP, CA, BMC and others at their own game, on their turf.  We’re trying to get to where the puck is going.   When there is a disruptive change in the market, new techniques can make entrants beat established giants.

So, we’ve done about $1B in M&A and R&D over the last few years – with the guiding principles of “what management functions, operational models will be important in a world dominated by new operational models that are “cloud like”?

These included acquisitions that focused on servers, clients, and applications – and up:

• Acquired Configuresoft – server (both ESX host level and guest OS level) configuration, control and compliance (including a great vCenter plugin) http://www.configuresoft.com/

 

Download high resolution in WMV or MOV format.

• Acquired FastScale – next generation guest image construction through on-demand composition http://www.fastscale.com/
• Acquired nLayers – very cool model-based (and no agents needed) application dependency mapping (including vCenter API integration, and vCollectors – which are virtual appliances) http://www.emc.com/products/family/smarts-family.htm.   That’s how we collect the info for the “EMC.com” portion of this demo of Data Center Insight here:

Download high resolution in WMV or MOV format.

• Acquired Infra – a very cool service desk-level application – something that ties all the application information back to business-level SLAs.

We also expanded out our assets that focus on the network, storage, and integrated infrastructure - and “down”

• Acquired SMARTS (http://www.emc.com/products/family/smarts-family.htm) this is a very extensive set of network management tools, focused on root cause, performance analytics and much, much more.   Widely used in the enterprise and service provider community.  Cisco OEMs this from EMC as Device Fault Manager (DFM).
• Acquired Voyence – Cisco Resells this product as Network Configuration Manager (NCM)
• Built Unified Infrastructure Manager (UIM) – this was a fork from Voyence - which specifically integrates Cisco UCS, Network, and Storage into template-oriented provisioning models.   You can see a demo of UIM v1 here, and I previewed UIMv2 at VMware Partner Exchange (read about that here)
  

  You can download the high-resolution version of this video here in MOV format.

• Built Data Center Insight (which actually pulls together all the info from the sources listed and more – like UCS manager).
• Continuously worked on Control Center (adding vCenter API integration even before it was cool to do that :-)

None of these are “perfect” products (there is no such thing, IMO), but they all fit into a framework of next-generation management – from the business SLA, to the application, all the way down through to the infrastructure.

What we started to realize on the EMC side is that all the “server and up” category management tools people started to ask for literally as part of the vCenter suite.   Likewise, if I thought about every time a customer has said one of the following when I was having a VMware-centric (not Ionix-Centric) conversation:

• “I really like the idea of ESX host profiles, but I wish it did more” (answer: SCM does that)
• “I wish I could have that compliance and remediation policy extend to the guest” (asnwer: SCM does that)
• “I wish guest composition wasn’t just from a template, but a ‘ingredients list’ from which i could make recipes on demand” (answer: Fastscale does that)
• “Appspeed is a cool idea, but it’s missing the idea of the fact that the vApps are all interdependent on each other and infrastructure – you need something to capture that relationship if you want to deliver SLA models” (answer: ADM does that)

For those of you wondering about a visual example of the power of this move…   At VMware Partner Exchange – remember the demo that Steve Herrod did of the future of VMware Studio, where they composed a developer’s vApp instance by dragging and dropping elements (kernel, app stack, database), which resulted in a custom-built (but easily and instantly built) vApp?   Fastscale can and will be a huge part of making that real and accelerating it.

Well… You get the picture.   

So – the server-and-up assets (SCM, FastScale, ADM, Infra) are now VMware assets.   EMC retains and focuses on the server-and-down assets (SMARTS, UIM, Voyence, UIM and Data Center Insight).

This move accelerates VMware – both with intellectual property, developers, and field resources in areas where they feel they could use them.

This move focuses EMC where we need to be focused – making sure that we provide the best, most integrated Information Infrastructure to support the Virtual Infrastructure domain, and transformational infrastructure management around VCE, Vblock, and integration with things like Redwood from VMware.

I’m pumped.  This is an example of the big moves we are doing and will continue to do to accelerate the Journey to the private cloud.

Comments

[Jose Ruelas]

this is great news!!!

kind regards
Jose R

[Jason Wicker]

Chad,

Killer way to lay this out. I hope you dont mind I shared this with the entire IONIX sales team.

BTW thanks again for the late night help with the Canadians in Boston :)

Jason

[Chad Sakac]

@jose - thanks!

@jason - please share - I hope it helps! clearly lots of change, but this has been thought through for a long time. To me (just one person's view), this is an obvious, natural move. Great for Ionix, great for VMware.

[Grant]

Wow thats the biggest load of FUD I have ever heard ! We definitely know who pays your salary !

So you couldn’t sell these products effectively as individual products (they were never best of breed , second string at best)

Then EMC acquired the products and couldn’t sell them , so rebranded them as Ionix and then still couldn’t sell them !
And have now handballed it to Vmware which will cripple them further with a Vmware only story .

Way to go on shooting yourselves in the foot !

[Stuart Savill]

Hey Chad - nice blog - been following for some time... and a heap of your stuff is "on the money", I gotta say - on this one i think people may be a little "non-believing" why? well...

I had an update on the Ionix product suite (And in this case - yes it was focused on the ECC side of the product, but was also looking at the more general approach) - and i was big time disappointed..

Firstly - EMC are still yet to get the base "nuts and bolts" bits right, and i am talking about all aspects of their own product set fully integrated into this suite in a TIMELY manner.

There are more innovative products breaking into (and already here!) the market that are now quicker to market, spread across more of the heterogeneous product set and more importantly ACTUALLY WORK & PERFORM (yes they may not be under one brand name - but they are out there).

I would also say (and i sort of agree with some of Grant's comments posted above) that this is no more than a re-branding exercise for loads of disparate parts and being able to put a bit more of a sales push on this stuff.

EMC have a lot to prove in the enterprise in these management spaces, once again i think that EMC may be biting off more than it can chew, and do a bunch of things badly, where it should be doing fewer things really well..

I for one - am a non-believer.....

Happy to have a chat about this...

Tweet - @stuiesav

[Chad Sakac]

@Grant - While everyone's entitled to their opinion, and I'm not going to censor folks, it's notable that you use a false identity (false email, an IP address from APJ, but don’t know with whom via ARIN). To continue with reasonable dialog, how about dropping the anonymity?

@Stu - Of all the products in the Ionix family, the one that disappoints me is Control Center. Customer feedback I hear consistently (and the product team knows) is that it's still to heavyweight, difficult to deploy and scale properly, and the "time to value" is really long. Some customers like it, many are frustrated.

The Control Center team is "back to the drawing board" working on their next major revision. There's also been a lot of dialog that the evolution of Storage Resource Management (SRM) as a product category has been bumpy. There are few customers super-happy about their tool (though I agree, there are some that they are generally happier about the one they are using than Control Center).

My personal opinion is that we have evolved that product to try to make a very small number of customers (the largest heterogenous storage customers in the world) happy, and they're still not happy. I wonder if we can ever make those 10 customers happy.

I think we should go back to the drawing board and start from first principles - and perhaps most importantly as opposed to trying to please the (perhaps?) unpleasable, make a simple product, that goes in fast and easy, can scale well (but not necessarily uber large) and delivers a low time-to-value for the storage customer looking for storage resource management.

Now, don't get me wrong, there are happy Control Center customers, and it still leads the SRM marketplace in share, but we want every product to bring joy and passion in customers - be the "best" in solving that problem. I don't think Control Center is there right now.

When I can say more about what we're doing about the future of SRM, I will - right now I don't know enough, but have been pushing the Control Center team for more.

What I **DO** know is that people talk to EMC about management **expect** to start that discussion about storage, and that's a place we have to nail. I almost wish that Control Center was not part of the Ionix family - as they're VERY different, but it IS part of that family, and we need to make it a great part of that family.

Conversely, customer feedback on Configuresoft (Server Configuration Manager), nLayers (SMART ADM), and Infra is consistently VERY positive. They go in and are installed in minutes in the case of SCM, and 1 day in the case of ADM. Time to value (really important in management land) is super-fast. Feedback on Fastscale is very positive, but we acquired them before they got big (very much acquired the tech, not the business) - the number of customers is relatively small.

Now, this (too me - and I do recognize that I have a bias - BUT I always only say what I believe based on everything I see, and try to avoid "self-drinking koolaid") reinforces my point:

- the set of new "startup acquired" things that were "not connected to infrastructure directly, but focused at the server, guest, and application" make more sense at VMware than at EMC.
VMware made their own choice on this one of what to take, and what not to take. The things that they got are FANTASTIC products. This will allow them to focus on integrating them into VMware’s management stack.

I know I'll never be able to convince an anonymous poster like "Grant", but there have been loads of occasions (and there will be more, including several very soon) where VMware does things that are actively in their interest and in a clear sense not in EMC's interest. There have been times where EMC has proposed things that VMware has out and out rejected. The board knows that VMware must be able to make the moves that make them continue to grow and succeed.

The flip-side (what this means to EMC) is that it means the Ionix team at EMC can focus on the products that ARE focused on infrastructure, including ones that are:
a) new "startup acquired, then home-built and need to evolve fast" (like Unified Infrastructure Manager) -
b) been part of the family for a while, products important to "non storage" parts of the portfolio focused on infrastructure (the broad network-centric parts of the smarts family that is OEMed to Cisco)
c) "home built, and needing a new foundation" (like Control Center)

I will say this - there are things that I see EMC do that are mistakes, and I've frequently blogged that they are that - a mistake. No one tells me what to say or do.

I agree that EMC where we don’t have industry-leading products that people love and are passionate about... Well, we need to focus, simplify and do further product integration and rationalization. Let's have a post at the end of 2010 on that topic - simplification and rationalization of the product breadth is furiously underway.

Back to the topic of this post... I genuinely believe that this is a very good move, and while I can see the perspective, can also say that it is not a "sales rebranding move".

And the other thing I would say is that ultimate customers make the choices, regardless of what I (or anyone else - including anonymous folks like "Grant") says.

[Martin G]

Great to hear the honest comments on CC; EMC have heard enough times from me about the subject. I feel your pain on CC's place in the Ionix family, it must be hard to have a product which is mostly loathed by it's user base (note I say mostly) and for many customers, it is the only exposure that they have to the Ionix product set...it may not be your flagship product but most customers will see it as that.

CC is one of those products which needs to be 'functionally matured' and then stripping right back to basics!

EMC have a whilst to go before they have credibility in the management space and it's now moving so fast, EMC have to run like heck to gain it.

[Stuart Savill]

Chad - Martin's post is on the money (as ever...) Credibility is everything in this space, it may be that the rest of the Ionix suite is the best thing since sliced bread, however - Most people's experience of EMC's management tool capability is ECC, therefore its what everyone uses to gauge EMC's competence level on management tools - therein lies the problem.

I will post no more on this topic - as I think we understand each other - but will call you later to discuss.

Cheers,

Stuart.

@stuiesav

[Terry]

None of these are “perfect” products (there is no such thing, IMO), but they all fit into a framework of next-generation management – from the business SLA, to the application, all the way down through to the infrastructure.

-----

I would dispute that the SMARTS product IS a best-of-breed product that is still unique and unlike no other in the market place, especially as you stated in the service provider community (COLT, Cable & Wireless, BT, Vodafone, Orange, SBC and Verizon) but most large enterprises run SMARTS as well.

SMARTS is such a unique technology that even Microsoft has included connectors to it.

Microsoft System Center Operations Manager 2007 and EMC Smarts Integration
http://www.emc.com/collateral/software/solution-overview/h4228-micro-smarts.pdf
http://www.microsoft.com/Presspass/press/2007/mar07/03-27EMCPR.mspx

[Chad Sakac]

@Terry - thanks for the comment, and I'm glad you like SMARTS and are passionate about it.

My comment was only to highlight that I'm not saying "RAH, RAH - our products are perfect!" but rather that many of these in the Ionix (and now VMware vCenter) family are really best-of-breed in their category.

Most customers still don't think of EMC naturally when the topic is "management", but we are indeed dead serious about the space.