« IBM and SUN OS Agreement - Announcement Today 8/16/07 | Home | Things that make me go Huh #004? »
My Views on OSS ESM (Part 3)
By John | August 16, 2007
Coincidentally, I had a conversation yesterday with two of the ten largest banks in the world. Both told me that 2008 looks to be a possible greenfield for some of their ESM initiatives, or, in other words, a do-over. Before I could even ask, they started talking about open source. One of the executives told me that anything is on the table and that some of their proprietary solutions are just not getting it done.
The press and the blogoshpere have recounted numerous conversations about how the big proprietary vendors are overpriced, too hard to implement, and vulnerable to certain open source products. Obviously, I am not the first to point out these flaws; however, for the first time, my specific customers are asking me about open source alternatives. In my little niche, I see some of my customers paying in excess of $10 million a year for ESM, and they are still not getting what they want.
To put things in perspective, most of the enterprise customers with whom I work fall into these 3 categories:
- A customer who never really gets the software to run, and the products never goes into production (i.e., shelfware). In most of these cases, the customer was sold or told that he or she didn’t need any extra resources (i.e., people) to manage the product, hence the inability to make the product work. Comically, I usually see this type of customer float back and forth between vendors every two or three years. I can’t tell you how many customers I have seen go from BMC to Tivoli then back to BMC and vice versa.
- A customer who buys the software through a re-seller and the re-seller brings in a lower cost consulting service to increase the seller’s margins. In these sites, the products are implemented, but the organization and the consultants never get to the root cause of why they installed the software. These customers usually live in a constant state of denial (i.e., virtual shelfware), thinking that the product is in production status but never receiving any real output from the products. In most of these cases, these customers basically run the product out of the box with few or no modifications, expending most of their efforts on keeping up the product’s infrastructure or applying maintenance. I once did a con call with such a customer, and I asked where all the event notifications were going. The answer: only to the customer’s own ESM group. Then I asked, “Are you telling me that no one in your organization gets any notifications from your ESM product?” I was told that the notifications were sent to the sys-admins when the product was first installed but that the sys-admins demanded that the notifications be turned off. Then I was given this minor correction: “Oh, wait a minute. We do have one”: the helpdesk gets a notification from ESM. “For what?” I asked. “They get a notification when the ESM product goes down.†So this customer had a product that gives no-value but, to add insult to injury, requires additional overhead on a helpdesk to manage it. You would be surprised to learn how many times I have had this conversation over the last decade.
- A customer who invests a significant resources into the vendor’s product to get the solution work. These customers almost always have at least two dedicated full time employees devoted to ESM (just the availability products). I have seen originations that have over 30 FTE’s dedicated to the base health and welfare of their IT infrastructure (i.e. ESM). Alternatively, the customer retains a staff of 3 to 5 full-time Tier 1 service professionals, each of whom averages around $200 dollars an hour. The most successful customer of this type has an ESM budget well over $10 million a year. Still, these customers do amazing things with their ESM products. In the end, they have customized the product so much that they have actually written their own product. To make things worse, their vendors constantly buy new companies, sometimes forcing these customers to upgrade their own in-house developed customizations to the newly acquired products (Version 2). When they try to get support, they are told by the vendor that they should not have customized the product in the first place. Interestingly enough, these customers are actually talking more and more about using open source for all the reasons listed. By the way, this includes a number of vendors, not just my favorite (e.g., BMC, Quest Software, HP and IBM).
The other problem most of my customers have with propriety vendors is that they are always selling marketectures. Every two years emerges a new buzz, on which all the products suddenly focus. The ironic thing is that, even though industry trends (i.e., the buzz) change every few years, the CODE doesn’t. In fact, almost all the proprietary vendors in ESM are running with code bases of over 10 years (even 20 in some cases). An amazing number of products have become ITIL compliant within the last 2 years even though ITIL has been around for over 10. Besides, products can never define process; only people can define process and then force products into that process. I was just commenting today that I stopped asking people what their SLA’s are because it’s too embarrassing. When I ask, “Do you have SLA’s?†they tell me, “Oh certainly.” If I question them further, asking what their SLA’s are, nine out of ten times they would respond, “Uptime.” If I ask any more questions, I may as well call these people idiots, which is not a good idea if you are in the services business. Even 10% would say silly things, like a disk’s used percent or high CPU. Come on, dudes, it’s almost 2008. SLA’s in 2008 have to start with business services and work their way down from there. You can not do SLA’s from a bottom up perspective.
Which brings me to my final point: If these customers have an opportunity for a do-over (i.e., a greenfield), then we need to make sure not to make the same mistakes all over again. Sure, open source is a fabulous enablement platform for going greenfiend, but maybe we should look at turning this square wheel into a round one at the same time. If you look at the whole industry debate of the Big Four vs. the Little Four, you see that most of the Little Four are just replicating the same mistakes of their forefathers, the Big Four. I have been scouring the internet compiling information for my customers and my community about open source solutions regarding ESM, and I am not finding things to be that different. In fact, the only major difference is the lexicon. I have been recently looking for an open source response time transaction tool (traditionally called Synthetic Transactions). I can’t tell you the number of times that I have been pointed to curl or wget as the solution. Even when I use terms like SOA to open source vendors, they tell me that they have redesigned their products to be SOA-based. I don’t even bother to tell them that I don’t give a rip how their code was designed, that all I care about is how they are going to monitor my customers’ SOA. I will agree that most of the new ESM-based open source tools are very agile and have extremely youthful code bases, which is good. I am not, however, hearing a different message. The industry has clearly tipped, and some big changes are coming. Let’s just hope that in 2010, we don’t have to do a greenfield.
Here are the links of my previous two articles on this subject:
Topics: 451, OSS, barcampesm, barcampnashville, bmc, caos, cmdb, esm, gartner, groundwork, groundworkopensource, hp, hyperic, ibm, itil, nagios, netcool, opennms, opensource, oscon, ossesm, tivoli, zabbix, zenoss |
August 17th, 2007 at 12:37 pm
[…] Posted by on 17 Aug 2007 at 05:36 pm | Tagged as: Enterprise Software, BMC, Opsware, CA, IBM, HP, EMC, Solarwinds Some mornings you’re just not inspired and you need something to jumpstart things. Today was one of those days. I was having writers block on what to blog for this week. So I used this as an opportunity to catch up on reading other blogs. Before I knew it mission accomplished, I found some amazing and refreshing perspectives on John Willis site that really got the mind racing. Specifically, it was his recent posting on Open Source Software (OSS) and Enterprise Systems Management (ESM) - My Views on OSS ESM (part 3). […]
August 17th, 2007 at 12:43 pm
John, great posting - thanks for the inspiration this morning for some of my own views on VC vs. ESM at http://42internetworks.com/blog/
August 21st, 2007 at 4:32 pm
[…] My Views on OSS ESM (Part 3) […]