This post originated from an RSS feed registered with Java Buzz
by Michael Cote.
Original Post: Enterprise Management Systems Solving New Problems
Feed Title: Cote's Weblog: Coding, Austin, etc.
Feed URL: https://cote.io/feed/
Feed Description: Using Java to get to the ideal state.
Cameron Sturdevant has an article in this week's eWeek on "Enterprise Management Systems," software that's used to monitor and manage computers and applications on an organization's network. For example, one of these management systems could be used to tell you your web site it down, preventing customers from buying your widgets, or that your email server is send out email very slowly, meaning that emails aren't getting sent out to your customers quickly enough.
It's not online yet (it'll probably show up in Sturdevant's column archive when it is), and it unfortunately doesn't mention BMC, but it has some interesting points:
The difficulty of installing and configuring management systems in the 90's has left companies with a bad feeling and, thus, a lack of trust and acceptance for these systems. Even nowadays, Sturdevant says, he's encountered horrid ease-of-use: one system he encountered had a "tome to support software deployment [that] was nearly 3 inches thick--and that was just for the pilot project." And, that's just for installing it, not using it.
Because the networking layer in most companies is more homogeneous nowadays (TCP/IP instead of a mix of token ring, IPX, NetWare, etc.), it's much easier to write effective management system tools: you don't have to support all those funky network protocols. Also, of course, the networks are faster.
Current tools are more flexible and component based, which makes "enterprise products much easier to integrate with third-party, best of breed tools." That is, among other things, that the systems management tools are easier to tailor fit to the needs of the business rather than the other way around.
The overall feel seems to be: do just what's needed with the minimal effort. This idea dovetails into the old "solve problems, don't provide features" mentality of product development.
In the context of "Enterprise Management Systems," what's slightly more complicated than that simple maxim is realizing that a successful product provides a platform that easily allows for solving new, unanticipated monitoring and management problems as they're encountered, preferably without having to upgrade the systems management software.
For example, if you've got a one of these products in place, and you add new hardware and applications to your shop, you want the product to be able to monitor and manage those new things without spending more money, and esp. man-hours, to upgrade or buy a new component: the ideal product should be able to solve new problems like this as they come along. Indeed, when larger monitoring and management products fails to do this, a lucrative market is created for the "best of breed" tools mentioned above, e.g., I've been told, NetIQ in the Windows world.
(As a plug, I think the product I work on, PATROL Express, currently does a pretty good job at the above.)
Previous Topic
