At the end of August, Brad scored a direct hit close to the bone.
He made the case that EMC hadn’t done the best job in the world in communicating what our overall strategy and direction, especially in light of all of the acquisitions.
Ouch! That hurt, because sitting inside EMC, it all seems perfectly obvious and sensible. And it made a lot of us think, yep he’s right, we probably have fallen a bit short in this regard.
So let’s see if I can get the basic ideas across in a short amount of real estate.
EMC thinks that information is an organization’s most important asset, and that information value is independent of any particular application, technology or use case.
Core management functions need to be implemented independent of any application, database, file system, operating system and so on. Doing it ad-hoc costs more, increases risks and keeps you from leveraging information in new ways.
Looked at that way, how many organizations have an independent strategy for managing their information? A game plan to lower the costs associated with owning it, reducing any risks, and leveraging its value? Not too many.
We believe that every information strategy will require an information infrastructure. Simply put, a set of shared information management services that store information intelligently, protects and secures it against risks, optimizes costs and service levels, and – ultimately – helps the organization leverage the value of its information in new ways.
information infrastructure =
shared services that store | protect |optimize | leverage.
So, when EMC says it’s an information infrastructure company, that’s what we’re trying to build. Recognize that information is its own asset, and help customers figure out a game plan with products and services to lower costs, reduce risks and find new value.
We don't think anyone else is well positioned in the industry to do this, and to do it in a way that paves the way for the next generation of IT thinking: SOA, grid, et. al.
So, if you're up for it, I'll take you through a brief tour of each of the layers, and you'll see where some of the recent acquisitions fit in.
Store:
Well, it’s our heritage. Depending on who’s counting, we’re either #1 or a close second in most categories. And there’s still work to do here: focusing on environmentals, data deduplication, simplification, ease of use. I don’t think we’ll ever run out of useful things to do here.
In an information infrastructure, you need to start with storing information intelligently. We think it's the foundation, because that's where information lives.
Protect:
Information has value, so it has to be protected against loss as well as unauthorized use.
Most of you are probably aware that EMC pretty much invented storage-based replication, and a significant number of large customers use products like SRDF and TimeFinder to protect important information.
We also think as the price of disk drops, more and more of recovery/archive information will want to be stored on very low cost disk, rather than tape. Tape doesn't go away, but its role is changing.
A few years back, we added Legato to the mix; it helped customers get to disk-based backup. We got into the VTL market in a big way.
And a recent acquisition (Kashya, now EMC RecoverPoint) brings CDP to the mix and does it out-of-band in the network. I try and stay away from product plugs in this blog, but if you're curious as to what next-gen remote replication looks like, go see RecoverPoint. Very cool.
All of this fits into our work around application consistency, integrating replication and recovery semantics with specific application environments (like Exchange!) and so on.
But there’s another, more recent wrinkle to this. A few years ago, we realized that most people were trying to secure infrastructure, not information, and needs would change. Yes, you can lock down the network, server, database, etc. -- but how do you lock down the information wherever it goes?
Think about emails, spreadsheets, memory sticks, etc. in a world where it's a crime to disclose sensitive customer information, and you'll get an appreciation of the scope of the problem.
We believed our customers would want an end-to-end approach: identity and access management at one end, and digital rights management at the other that followed the information wherever it went. That’s largely the rationale behind the RSA and Authentica acquisitions.
Optimize:
Lots to optimize in IT. The bigger your shop is, the more you care about optimization. We’ve picked three areas of optimization to focus on: infrastructure, information management, and service level delivery.
Our big play for infrastructure optimization is virtualization: servers, files and storage. VMware for servers, RainFinity for files and Invista for storage. Most people are attracted to higher utilization rates that they see with virtualization, but the bigger payoff seems to be flexibility: the ability to move things around without downtime. Create a capability to move an application and its information from A to B non-disruptively, and IT has a whole new world in terms of being responsive to the business.
We’ve also invested heavily in tools that can pick up a piece of information and figure out what to do with it. We’ve got our Xtender family that can deal with email, files and databases. Documentum has a neat capability called Content Storage Services that uses metadata to automatically move information to different service levels, decide how long it needs to be kept, and so on. And there’s more coming here soon that I can’t quite talk about yet.
But the big idea here is using information to tell the infrastructure what to do. Imagine FedEx trying to run a package delivery business without labels on the package. That's a rough analogy of what IT professionals try to do with information: manage it without labels. We'd like to fix that problem.
The final area we think is important is service level delivery. The IT stack is getting more complex, many businesses are contemplating SOA, and the more distributed and decomposed application stacks become, the harder it gets to manage end-to-end service level delivery. End users complain loudly when service levels drop and IT can’t figure out where the problem is. And we think it will get worse before it gets better.
We want to build out three important capabilities here. First, real-time discovery of applications, infrastructure and relationships. Second, populate a powerful, open model that focuses on relationships and interactions. And finally, build applications that can use the model to do real-time root cause analysis if there’s a service level issue. We acquired nLayers to do the first, and Smarts to do the model and the apps. More work to do here, but it’s a promising start.
Leverage:
All that great information can give back value to the business. Take everything you know about customers, products, markets, etc. and put it to work in new ways. The problem is that all this information now is rich content, and not tidy transactions sitting in a data warehouse.
We thought we needed a capability that could do this, and focused on the newer challenges around content, collaboration and business process management. That’s where Documentum and related acquisitions fit in.
Finally, you need the wrapper. Solution templates that help customers implement quickly. Services that range from consulting to change management. Productive partnerships with others who bring value to customers.
So, in a nutshell, that’s what EMC is doing: information infrastructure. Shared information management services that store, protect, optimize and leverage information, which -- if you think about it -- is what IT should really be all about.
If you do it right, you help customers save money, reduce risks and become more responsive to the business. It's a big thought, to be sure.
If it still doesn’t make sense, please blame me and my erratic communication skills ... thanks!
Comments