This paper discusses and explores planning, requirements gathering,
Vendor selection, product procurement, designs, deployment and support
issues involving
IT project initiatives that use software commonly referred to as
"Enterprise Monitoring Software". This would include, but not be
limited to server and network monitoring, job control, asset management,
software distribution tools, backup and disastery recovery, trouble
ticket systems, performance monitoring and much more.
Simple question. But one that some IT Managers give little deep thought
to. Many IT Managers came up through the ranks because of their ability
to put out big fires. Others were promoted because they built
data centers that were robust and impervious to fires. Doing
what you've been rewarded for in the past will obviously have a strong
influence on one's management philosophy moving forward.
Obviously, the prime objective is to keep the network humming along.
But how is this done? By spending the least possible amount of money
using free or "cut rate" products? Or does your company IT focus on
buying the best it can afford to obtain enterprise reliability, up time
and customer happiness?
Does your company look for quick, inexpensive fixes for problems?
Or does your company insist on solutions that will permanently solve problems?
Does your organization
believe that spending money is the solution to all problems?
Does price, cost and/or "political" motivations drive IT decision
making?
Does your IT department operate using a reactive or proactive model?
Does IT seek
knowledge and assistance from professional consultants with specialized
skills to move
the organization upwards on the ITSM model? Or does every new
initiative become just another task for an already over worked
IT Staff? Please understand that there are no "wrong" answers to any of these
questions. But in
my opinion, making sensible decisions becomes a lot easier once
managers fully consider what the company IT management objectives and
methodology are.
As you probably know, when you get up into these price ranges,
monitoring software vendors are
not always meticulously forthcoming when it comes to providing reliable and
accurate information. I happen to be one of a very few in the whole wide
world who offers a consulting service who does NOT sell software. I have
absolutely no financial stake in the product purchase decision which makes
me a very strong customer advocate.
I work for you.
The HP, BMC, IBM, CA and so on guys you are talking to do not
have your best interests in mind when they provide you with "information"
and "advice". All they care about is getting that fat commission. This is
a given.
As you know, spending huge sums of money on software that turns out to be a
disaster can be a very career limiting mistake. This is a big motivation
for why smart IT managers engage me to help design and build their NOC.
With a Home Depot on every corner, America has become a Country of
"do it yourself-ers". Fine if all you want to do is fix a leaky
faucet. Risky if you're planning on repairing your home's
foundation.
"Free" software is worth every penny you paid for it. And bear in mind that
the Employee(s) you "volunteered" to investigate, build and support this
"free" software aren't working for free. Keeping a NOC running is a full
time job! If you happen to have an experienced $80+k per year UNIX Engineer
just sitting around doing nothing, then by all means... building a NOC using
"free" open source software may work out for you.
Things to consider though... NOCs built using "free" open source software
tend to be HIGHLY customized. This means that if the guy who built it
leaves the company, dies or moves on to a different assignment, chances
are slim that someone else is going to be able to just seamlessly slip
in and take over. One of the things you get with expensive vendor software
is standardization. Pretty much any experienced Openview guy can take over
an Openview environment that was built by someone else. This is definitely
not true of "free" products such as Nagios.
Another important consideration regarding "free" software are it's
capabilities. It is reasonable to assume that a product that sells for
$100+k and is supported by a legion of VARs and Vendor Support folks might
have one or two features that "free" open source software lacks. This is
at least in part due to the fact that most open source projects are created
by talented programmers with too much free time on their hands. These
products are almost always developed as Engineering driven initiatives....
creating "cool" stuff, versus Marketing driven, profit motivated products
that seek to fill specific Customer needs.
I operate my own professional NOC.
I have been working with network and server monitoring software for the
past 30 years
so I am not exactly a novice with this technology. I certainly could have
saved a lot of money using "free" software. And I did evalute a few of the
top performers in the open source area before I went into the
Monitoring Service Provider (MSP)
business. But even the best of the "free"
stuff did not come even close to the kind of monitoring capabilities I
needed so I went with professional grade Vendor software instead. In my
opinion, there is just no comparison between software like OpenNMS, Nagios
and so on as compared to widely used and globally accepted products such as HP
Openview Network Node Manager (NNM). And NNM is Openview's entry level
product! HP Openview Operations (OVO), and to a somewhat lesser degree,
BMC Patrol, sets the standard
by which all other monitoring software products are measured.
"Free" open source software is certainly better than nothing. But if you
are serious about moving your IT environment out of level zero of the
ITSM model
(chaos mode), "free" software is not the way to get there.
You should be looking at industry accepted, industrial strength,
proactive monitoring products that are easy to use and that integrate
into a seamless enterprise solution... not some cobbled together, Mickey
Mouse heap of "free" stuff.
Home-based in Portland, Oregon but I will go anywhere, any time, to design
and build your HP Openview, BMC Patrol,
IBM Tivoli, CA Computer Associates Unicenter, Micromuse, Microsoft MOM,
SunNet Manager, Optivity, etc. Network Operations Center (NOC).
More NOC design and implementation details can be found here.
Now that you've found me, why not bookmark this page or e-mail me right away!
I can save you a pile of bucks and help you avoid a lot of pain by guiding
you towards making sensible, appropriate and realistic software purchase
decisions. It's not uncommon for IT Managers to spend tens of thousands of
dollars on software that never gets deployed or is only marginally deployed.
I can assure you that my rates are probably a lot less expensive than what
your company paid for all of the shelfware you happen to have laying around.
You say you've already purchased software and you need someone to turn it
up? I can save you money there as well! Most of the Technicians a VAR
will charge you $200 per hour to have on your site know little more than
how to install software and do basic configuration work. The last time
they built an enterprise class NOC was never. And of course their
motivation is going to be to rack up as many billable hours as possible
while trying to sell you even more software... most of which you don't
need anyway. In contrast, I've built many, many NOCs over the past several
decades and my motivation is to build you a NOC you are really going to
love. And a NOC that your BOSSES are going to love! If I make YOU happy,
you are probably going to want me to come back from time to time and
developing long term relationships are what I am all about. VARs just
want to sell you software and the chances are, if you happen to need
more consulting help, the guy the VAR put on your site for all the
money in the world won't even be working there next time you need someone.
Over 40 years of computer, networking, electronics and broadcast radio
experience. Extensive exposure to customers in a field engineering
and IT environment. Solid, proven project management background
as well as headcount and P&L management experience. Heavily
traveled worldwide to resolve technical problems and customer issues.
My most recent work has been in a network and systems engineering capacity,
designing, implementing and managing Remote Management Systems and
Network Operations Centers (NOC). I have over 30 years of experience
managing UNIX (Sun Solaris, Linux, IBM AIX, HP-UX, etc.) servers and
networking infrastructure in a high availability, production
environment.
I was born with a soldering iron in my hand.....
My systems support experience pre-dates the creation of the full screen editor,
Microsoft Corporation, Windows and DOS as well as the invention of the
microprocessor..... and the disk drive, just to name a few.
My UNIX experience pre-dates the creation of Sun Microsystems.
My networking experience predates the founding of Cisco Systems, the
creation of the DARPA Internet, web
browsers, DNS, bridges, routers, ethernet, the 1200 baud modem and most of
the common capabilities that we take for granted today.
Created the concept of
Monitoring Service Provider
(MSP) and the business
model of outsourced IT service availability monitoring utilizing a cost center NOC that I designed and turned into a highly profitable revenue center.
Designed and implemented some of the early SNMP remote management
systems used on the West Coast. Attained Developer Architect level
skills with SunNet Manager, Sun Solstice Enterprise Manager, IBM Tivoli
Netview for RS6000, Bay Networks Optivity, HP
Openview Operations, Openview NNM Network Node Manager, Network General
Distributed Sniffers, Network General SLM, Microsoft MOM and
SMS and more. Have designed, implemented and managed many
carrier enterprise class 7x24x365 Remote management facilities and Network
Operations Centers in high performance SLA environments.
Developed, prototyped and implemented
remote diagnostic troubleshooting capabilities
for Digital Equipment Corporation (D.E.C.) PDP-11 and VAX VMS customers
(including satisfying security issues for DOD/DOE) in the late 1970's,
before the technology was proven or popular.
