Network World's Network/Systems Management Newsletter, 06/19/06
EMC's purchase of nLayers helps EMC unite ILM with true enterprise mgmt. systems
By Dennis Drogseth

Two weeks ago, I wrote about mergers and acquisitions in the network management space and last week I wrote about HP's OEM deal with Voyence. And this week I'm writing about EMC's announced acquisition of nLayers. The first thing to say was that this wasn't premeditated. Each week I had intended to write a follow-on to the CMDB report findings, but I am waiting for the finished report to get through production first. So in other words - it's been pure happenstance!

That, of course, makes the point all the more strongly that strategic mergers, acquisitions, and OEM actions are probably the single most obvious force transforming the face of the IT management marketplace today. And both last week's and this week's news (as I'm writing vs. when you read this) - show how modest moves from a business investment perspective could deliver huge potential upside if the right strategic planning and commitments are made. They are both symptomatic of a new marketplace in which architecture, design, cross-domain analytics and managing change more proactively are gradually replacing the old, lame, functional checklist SWOTs (strengths, weaknesses, opportunities, threats) analysis that still hang around like pterodactyls squinting dimly at a Holocene zoo.

NLayers participates in a market most commonly called "application dependency mapping." Whether that is truly a "market" or not - and I am beginning to feel that the most interesting areas in IT management overall are not strictly speaking "markets" - it is at the vortex of the huge shifts that are transforming current solutions into next-generation capabilities.

Applications such as nLayers provide insight into how infrastructure components map dynamically to application services, and how configuration changes across the infrastructure may or may not impact application performance. Some of the other vendors in this market include Cendura, mValent, Tideway and Troux as independents; and Mercury (which acquired Appilog), IBM (which acquired Collation), Symantec (which acquired Relicore) and Opsware (which just announced its own capability). HP is similarly about to introduce an application dependency mapping solution.

The solutions vary in terms of how real-time they are (nLayers being at the high-end of true "real-timeness"), how detailed they are in capturing infrastructure configurations (typically these solutions focus more on systems rather than in-depth network configuration changes), and whether they capture application volumes or not - an area in which nLayers excels. As a class, they have been probably the single most evident catalysts for another even grander catalyst - the CMDB itself. I think it's safe to say that the stunning rise in plans for CMDB adoptions has been triggered in part by the fact that application dependency mapping is coming of age - bringing a sense that at least there's light at the end of the tunnel in achieving ITIL's seemingly impossible dream of correlating services to infrastructure components in complex and dynamic environments.

The acquisition announcement made by EMC on June 7 follows an OEM deal between EMC Smarts and nLayers that was announced at Interop about a month before - focused on what EMA would call a more dynamic approach to "cross-domain correlation." The value here is in teaming up one of the industry's most effective capabilities for gathering and modeling information about application-to-infrastructure "fingerprints" (using an nLayers term) with one of the industry's most proven analytic engines - EMC Smarts. Thinking structurally (and trust me it's becoming increasingly the only way to think!) this makes huge sense. So at minimum, the acquisition should help to strengthen that integration.

But nLayers also provides strong enabling capabilities for CMDB adoptions - and in a number of environments has effectively seeded CMDBs. Since I haven't had a chance to write that follow-on column on the "two CMDBs" yet - I'll have to say here, briefly, that EMA has observed fundamentally two distinct classes of CMDBs emerging in the market. One is focused on change impact management, is process-centric, and hence built around process control systems with robust inventory and configuration inputs. This is the CMDB that ITIL would most likely recognize as its own. The second CMDB class is effectively real-time and focused on real-time service assurance. Its focus then is service impact management and typically involves reconciliation systems that don't actually move data, but access is based on policy - across a set of pre-identified data sources. Most application dependency players support the former type of CMDB.

But nLayers plays to both, and its immediate integrations with EMC Smarts favor the latter type. By the way, complete CMDB systems can include both flavors - a discussion I'll examine the upcoming column as a follow up to the research.

So as I'm fond of saying about this acquisition - the "elephant in the living room" is the chance for EMC to open up directions in the area of a CMDB. EMC has recently made statements about its support for service-oriented architectures, which might seem to favor this opportunity, as well. But for the CMDB direction to become optimized, EMC will need to make further technology investments and evolve its management portfolio to become an intelligent and viable center for multi-brand integration. This is of course a cultural as well as technology challenge, and nine out of 10 vendors fail the cultural transformation test. Yet, in spite of the obstacles, moving in this direction could help more than any other single thing to take EMC further beyond storage per se, uniting information lifecycle management with the even larger arena of true enterprise management systems.

Copyright Network World, Inc., 2006

Questions or problems regarding this web site should be directed to abeckman@outdoorssite.com.

Copyright 2008 Art Beckman. All rights reserved.

Last Modified: March 9, 2008