* Why CA’s eHealth 6.0 is on track to provide integrated network and applications management
In a recent newsletter, I described how systems management platform vendors could compete in the network management space. I mentioned that there are a number of things to look for in assessing your system management vendor’s progress, including the commitment to integrate network and applications management, particularly in terms of performance as well as availability.
Is your vendor reinvesting in flow-based management, is it committed to configuration management and integrated service-level management (after all SLM began in the network operations center) and is it leveraging such capabilities as analytics, discovery and inventory as building blocks rather than as siloed functions targeted at an increasingly fuzzy network management market? If you’re getting your network management from a platform vendor, you should look for more creative points of integration with the rest of IT service management (ITSM) or business service management.
In a bit of coincidental timing, days after writing the newsletter cited above, CA briefed me on eHealth 6.0, CA’s systems and applications management product, which it inherited through its 2005 acquisition of Concord. The product is expected to be available on Oct. 30.
Some of the more significant features of eHealth 6.0 include:
* Improved eHealth reporting functionality through its OEM partnership with business intelligence vendor Cognos. Through this and its own developments, CA will enable users to create a hierarchy of customized reports and to search for linkages across information using a drag-and-drop interface.
* A single user interface for administration.
* Improved integration with Spectrum and access to data in Spectrum’s database, which provides information on availability, topological awareness and service impact.
* An improved database API for third-party sources. (CA has partnerships with vendors such as Fluke for packet analysis, Opnet for capacity planning and optimization, Splunk for detailed syslog data and Voyence for network configuration management.)
* Poll and store data at 30 second intervals. This topic bears more discussion than I can provide here, but it’s clearly a move towards improved time granularity.
* Improvements in NetFlow-based performance management for traffic volumes, and additional traffic accountant reports.
Longer term, CA will address much of the need to approach network management technologies as integrated building blocks for a broader service management strategy leveraging core, extensible technologies (analytics, discovery, CMDB, etc.).
The line up looks good from my own set of criteria as documented in the recent newsletter. However, the “devil is in the details” and while all these directions seem to be well in line, CA’s success will depend on product execution, including ease of deployment, cost and whether CA will continue with its commitment to “modularity” for choice and flexibility. CA will also have to continue to execute on its “cultural revolution” in how it interfaces with its customers and how sales gets its arms around modular solution sells aligned to business processes vs. selling software “things” to manage infrastructure “things.”
One final comment, which is somewhat orthogonal but which came out of our CA briefing. In response to a few of my questions, CA mentioned that its customers’ priorities divided roughly along the following lines: Fortune 500 enterprises were still largely looking to buy “network management for the NOC” as a separate organizational sell, while mid-tier enterprises were generally more interested in integrated network systems and applications management. This trend, while not universal, parallels much of EMA’s observations of CMDB deployments in large organizations. Such companies are embracing CMDB but have politically-entrenched subgroups that must first be integrated.
New models for ITSM, in which the infrastructure is managed in a collaborative way to support application business services, may evolve first in the mid-tier. This should be something of a wake-up call to vendors expecting that early adopters will continue to be giant (and albeit rich) IT groups in politically charged organizations. (Think of, for instance, the failure of the larger carriers to even begin to come alive to the need for more progressive cross-silo services and management except in the last few years.) If this is true, and it seems to be, progressive solutions will have to be priced and designed as much for mid-tier adoptions as for the high-end.
* CORRECTION: The URL to EMA’s Autonomic Computing survey, as mentioned in last Wednesday’s Network/Systems Management Newsletter, was incorrect. To take part in the survey, please go here. We apologize for the confusion.
