Don't Use a Discovery Tool as a Discovery Toolset. Don't.

 

My message for today: Don't try to use a discovery tool as a discovery toolset.  There seems to be a lot of consternation over function and capabilities between discovery technology.  As much as we like to say discovery is commodotized these days, there still seems to remain some distinct competetive positioning and evolution for this space.  Let's build a map!

 

JR's Discovery Technology Triant:

discovery triant.jpg 

 

A tri-what?  It's not quite a quadrant, so I'll call it one less.  The distinguishing characteristics:  You got your tools on the left, you got your tool sets on the right.  You got your agentless technology more towards the top, agents more towards the bottom.  But the funny thing, towards the left, there's a blending, a graying, of the space between the tools.

 

Tools:   The tools are over to the left.  They occupy generally less space than the quandrant-like "spaces" on the right.  They're generally more limited - no wait, let's say "focused" in function.  Maybe they specialize in a particular protocol such as SNMP.  Maybe they're part of a solution that requires special-purpose discovery, for example, for network management or operations monitoring.  They don't generally have a ton of configurable or expandable options. 

 

Toolsets:  The toolsets on the right fall into two broad categories, agent-based and agentless.  There are other ways to categorize toolsets but this one worked for me. (tell me if it works or not for you).   I used agent vs. agentless for a number of reasons, for example, HP discovery technology resembles the map.   But the most important distinction is dependency discovery.  Dependencies are required to fulfill certain use cases, such as, impact analysis, service modeling, data center transformations, application mapping, and many service transition-type use cases.  Dependency discovery is a primary characteristics of the type of agentless toolset like HP DDMA.

 

Back to the moral of the story:  Have you ever tried to use a tool as a toolset?  Did you succeed?  If you did, I'd bet you had some help, like a fairly small shop or a lot of organizational control.  Big, heterogeneous data centers  cannot be adequately discovered using only a tool.  Yes, you can get all the devices.  Yes, you can patch stuff up and inventory everything.  Yes, you can even use an agentless protocol like ICMP and claim agentless capabilities.  But that's like the old marketing hype:  "This device has wireless printing capabilities, but you still can't quite actually print from it without spending more money on something else."

 

Conclusion:   Don't try to use a discovery tool as a discovery toolset in a non-trivial environment.  You'll hurt all over.  You'll save a dime and then do some time, in discovery jail.  What looks like a good idea on the front side of a project will turn out to look like a dumb decision on the back side.  If all the customers I've talked to are right.  If you are an exception, we'd love to hear from you.

 

Thanks for watching! 

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the Community Guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
About the Author
Jody Roberts is a researcher, author, and customer advocate in the Product Foundation Services (PFS) group in HP Software. Jody has worked ...


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation