My Friday Rant -- Integration

Actually, this week it turns out people are ranting at me, rather than the other way around.

By all accounts, it was a great week: did a great broadcast thing on Tuesday, met with a great customer/partner on Wed, and had a similar experience on Thursday.

Along the way, I wrote a short post about my early days at EMC -- one of those intensely personal experiences I felt like sharing. 

The message was simple: don't rest on your laurels, because things change fast.

Some people saw it for what it was -- just me sharing my personal experiences and perspectives.  Others saw it as an attack on some of EMC's competitors.  Either way, it's a message that bears repeating.

But, somewhere in the ranting, a few points were made that deserve some exploration.

W. Curtis Preston Comments

One of the most interesting comments came from Curtis, whose basic thesis (once you look past the vinegar) was basically Where's The Integration? 

Since it was late at night, I made the mistake of offering a joking comment in response, which he took as a personal insult, a nd .. well ... we got off-topic a bit.  Sorry, Curtis.

So, in an effort to comeback to a worthwhile discussion, let's address the underlying strategic question: EMC has made all these acquisitions (I think we're over 50 or so), where's the integration, and -- ultimately -- the value for customers?

Since I'm so close to all of this, I see it everywhere, but not everyone has the benefit of my perspective -- so let's see what we can do here.

A Framework For An Integration Discussion

Consider any single product.  By definition, it should be reasonably integrated not only within its internal function, but its interactions with external entities that matter -- wifi networks, for example.

Now, add a second product.  It's pretty easy to see where the integration should be done to make them act as one.  You could probably extend this model to the third product, and maybe the fourth.

But that sort of simplistic integration model utterly breaks down when you consider literally dozens and dozens of products that span a broad swath of the IT domain -- from storage to security to virtualization to service delivery management to content management to home backup to ... well, you get the idea.

The notion of "everything has to tightly and seamlessly integrate with everything else in the portfolio, including everything that customers use from other vendors" has to evolve a bit.

First, it's not practical to do an "everything integrates with everything in every possible way" scenario -- it's a problem in combinatorics.  Second, real IT users care more about solving problems, rather than arbitrarily complex integration matrices.  And, finally, there's only so much R+D money available to do integration, and it's best to spend it where it creates the most value.

So, let me give you my view on how we're approaching this whole "integration" question as a basis for further discussion.

Strategic Integration Themes

At the top level, there are a set of cross-architectural integration themes that you'll find at EMC.  These are broad, strategic goals that are backed up by literally dozens and dozens of explicit integration points.

Probably the most visible one these days is private cloud 

This one is about integrating technology from three companies (four if you count Intel!), services and a broad ecosystem.  Whether it's all the work we're doing to integrate around virtual machines, or how we express these concepts in easy-to-consume packages like a Vblock, you can see it everywhere when you start looking.

If you take a walk through storage, management, data protection and security, you'll find several dozen explicit integration points, some more visible that others, like EMC Ionix UIM.  Extend that to EMC Proven Solutions, EMC Services, into VCE and Acadia, and on to our partners -- most of the value comes from the integration, rather than the individual point technologies.

My personal focus as of late has been on the service provider side of private clouds, and that's a key part of creating an integrated ecosystem.

The next strong sub-theme is information management

Whether it's autotiering with FAST, integrating backup and archiving, storage platforms that support rich metadata (Centera and Atmos), applications that can populate and exploit that metadata (Documentum, SourceOne, Kazeon, etc.) -- go looking for integration around information management, and you'll find dozens and dozens of cool integration points.

We're all looking to manage our information as effectively as we've learned to manage money.  We need better approaches to balancing cost, risk and value.  Some of that has to be done without having any explicit knowledge about how the information was created and will be used -- consider FAST as an example of this.

Other times, we've got the ability to use metadata to do a better job at managing information.  Consider Documentem, SourceOne and Kazeon's ability to generate and exploit metadata, either using traditional storage, or more advanced storage platforms that understand metadata, objects and policies.

And there's an end-to-end ILM-ish theme at play here -- from primary to secondary to backup to archive to retention -- witness all the integration that's happened around Networker, Avamar, DataDomain, Data Protection Advisor, move to Atmos clouds, etc.

And I'd offer that the final -- and perhaps most important -- strategic sub-theme is integrated information security. 

Since our acquisition of RSA, we've infused security DNA and technology into everything we do.  We build, integrated and deliver our capabilities in a far more secure manner than we were ever able to do so before.

Take, as an example, the integrated security capabilities on VMAX, for example.  Or integrated key management as a part of storage encryption.  Or how we use RSA's SEIM capabilities (enVision) to enhance what Ionix does, and vice versa.  Or perhaps how we tie RSA's DLP capabilities to bring in eDiscovery workflows.

Stepping back a bit, look how integrated information security plays with the first two themes.  Nobody is going to trust any form of cloud -- public or private -- unless it's provably secure, and better than what it replaces.  That's rapidly becoming the case.

Or how a key part of information management involves identifying and securing critical information assets.  Bottom line, there's strong interplay between these strategic integration themes that needs to be understood in context.

Disclaimer #1: much more to do and our work is never done.  And maybe we're not doing as good a job as we should be promoting these strategic integration themes.

Disclaimer #2: there are a few more "strategic integration themes" at EMC, I've just given you some of the more visible ones to consider.

Solution Integration Themes

Most people tend to look at individual product technologies in terms of feature/function and certain aspects of integration.

Well, those continue to be important, but the importance of integrated solutioneering has risen dramatically in the last few year.  Providing customers and partners with integrated solutions that are pre-qualified and supported as a whole creates enormous value by accelerating time-to-value and reducing certain forms of risk for IT practitioners.

EMC has invested an enormous amount of time and effort creating EMC Proven Solutions around dozens and dozens of real-world use cases -- from common applications (SAP, Oracle, Microsoft) to fully virtualized environments and much, much more.

If you haven't taken a tour through the portfolio of EMC Proven Solutions lately, you should.  You'll be pleasantly surprised with the breadth and the depth of the integration work they've done, and continue to do.  I think we're getting pretty good at it, and our customers and partners tell us that it's a good thing.

In some sense, EMC Proven Solutions is nothing more than a conceptual extension to EMC's renowned E-Lab.  The pitch is simple: we spend the time so you don't have to.

Product Integration Themes

As you look at EMC's portfolio, there are distinct product clusters.  Storage, for example.  Backup and archiving, for another.  Ionix and the IIG portfolio, and RSA, and ...

Each of these product groups has an agenda for integrating across their respective portfolios.  For example, I think that the BRS group is doing a good job towards integrating Avamar, DataDomain, DPA, Networker, et. al. and some useful things are popping out, like BOOST.

There's also cross-product group integration.  Generally speaking, things are managed by Ionix, secured by RSA, backed up by BRS, workflowed and content managed by IIG -- and virtualized by VMware.

Although VMware is a completely separate entity from EMC, it is the undisputed target of much our integration work.  Witness the vast array of storage integrations, security integrations, management integrations, backup integrations, etc.

That's relevant not only in a tactical sense (people use a lot of VMware these days!) but in a strategic sense as well -- think private cloud, VCE and Vblock.

The Bottom Line

I think Curtis' comment provoked an interesting discussion.

From my perspective there's a veritable river of integration coming out of EMC these days, if you know where to look.  As a small example, here's just a few of the announcements from the past 30 days along these themes.

Here, here, here, here, here, here, here, here, here, here, here, here  -- there's more, but I think you get the point.  Each reflects a strong "integration" theme of one kind or another.

But there are important questions that need to be discussed.

First, does integration matter? 

The answer is clearly -- yes -- and its importance is growing over time as complexity grows, and greater simplicity is demanded.

Second, what matters more -- integration between products from a single vendor, integration between those products and the outside world, or integration around solving customer problems? 

Posed this way, the answer is obvious -- integration between vendor and non-vendor products that solve real-world challenges

What's really interesting is the third question -- should the focus of integration be around solving today's problems, preparing for the future, or perhaps a bit of both?

And the answer to that one I'll leave to you ...

Comments

[W. Curtis Preston]

Chuck,

What vinegar? Seriously. My initial reply to you was about as bland as can be! I listed a brief history of product acquisitions by your company, then listed a similar brief history of another company that I did not name. Then I said some will like one story and others will like the other story. Where exactly is the vinegar in that?

As to my "basic thesis," as you call it. It was this (quoted from my comment":

"But the main thing I'm trying to say with my comment is that the veiled implications of this article do not apply to the company that you're trying to apply them to. Neither them, their culture, nor their product line is anything like the EMC of old."

And I stand by that comment. I believed then (and now) that your previous post was a veiled attempt at slamming NetApp. But I also believe that NetApp now is nothing like the EMC of old, so therefore do not agree with the main point of your post.

And seriously. Next time you try a "joke?" Try putting some smilies or something. Seriously. And be even more careful with "insult" jokes. You're not Don Rickles, dude. And you and I don't know each other that well.

[Jay Ryan]

I don't think it's just a question of having the products actually integrated, but rather that they are SOLD as if they are seamlessly integrated. This pitch is all made under the "ONE" mantra. The problem comes in when the products are attempted to be implemented and you encounter an "opportunity". Everyone just starts pointing fingers and blame any issues on the fact that there are internal struggles between the aquired organizations. You don't have a single point of resolution and the sales guy has "moved on".