Showing results for 
Search instead for 
Did you mean: 

What's most important to you when choosing an NMS?

Level 11

I’ve worked with a few different network management systems in my career, some commercial and some open source. Based on my experience with each one of them, I’ve developed certain qualities that I look for when deciding on which product to recommend or use.

One common theme that always seems to come up when comparing experiences with my peers is how easy it is to implement and operate product $A vs product $B?

In my opinion, implementation and operation of a product are critical when accessing any product, not just a NMS. If it’s not easy to implement, how are you ever going to get it off the ground? Will training be necessary just to install it? IF you ever do get it off the ground and running, will it take a small army to keep it going? Will you have to dedicate time and resources each day just to cultivate that product? How can you trust a product with the management and monitoring of your network environment, if it crashes all the time or if you have to be a Jedi to unlock all of its mystical bells and whistles?

With that being said, what do you look for in a network management system? Easy to install? Intuitive interface? All the features you could wish for? Is cost the ultimate factor? I’d love to hear what you all think.



Level 9

A couple of things:

  • I have no time to aggregate data sets myself. That's why I buy a product. So at a base level, the product should produce graphs, reports & tables that are superior or equal to what I can build in Excel. That means pivot tables, or something similar.
    • Nice to hve: During #VFD4 I remarked that it'd be neat if I could query the dataset myself; select net_counters from *day* where latency > 30 left_join helpdesk_email_counts or whatever. We all know it's SQL on the back-end, expose that power to me on the front!
  • When I want to dis-aggregate the data, the product should not fight me
  • I believe in collecting counters, flows, metrics and timers across the whole stack; that's empirical data I can use. So the product ideally is flexible enough to hoover up all my SNMP traps, my events, my syslogs, even my AD replication tests (I can run repadmin /latency, why can't the NMS?)
  • Most systems keep 2 weeks of data, or however much you can afford on storage. That's a dumb approach. Keep the two weeks by default, but NMS should intelligently baseline things, and keep flow data from periods when the baseline was exceeded by a given percent.
    • Gamechanger: NMS texts me and reminds ME that month end is coming, that it's worried about load on the filer because disk queue length is trending 7% higher than this time last month.
  • Deployment-wise, I favor active directory & group policy. Please package your agent into an .msi file that i can modify & install silently, or package into a task sequence. Please no Java runtime dependency.
  • Any NMS system that doesn't offer me the ability to secure it via a SAN cert or wildcard cert is probably not going to get installed;
    • It'd also be neat if documentation & support personnel nagged customers to use FQDN, not ip addresses, to access their NMS. I will have good hygiene and practice in my environment so help me $deity.

Finally: monitoring as a term feels too reactive, old-fashioned. I have worked in environments where the expectation is that I will know about resource contention before it happens. To get a text 30 seconds, or 30 minutes after resource contention impacts the business is too late. User & business don't care that NMS caught the problem after the fact.      I need predictive analysis, not reactive reporting.

Level 16

When i look at NMS software that what  importent to me:

Usability&Flexibility - NMS is there to help the users "navigate in a multi-vendor soup".

Readability -If you can't trust the data /performance/ or NMS vendor support -or big problems 🙂

TCO -easy to predict your cost /licenses /support.


Level 9

When looking at NMS solutions I look for:

1. Ease of implementation

2. Amount of training required by staff to utilize

3. Reliability of NMS solution

4. Cost

Level 15
  1. Reliability of Solution
  2. Implementation
  3. Security
  4. Technical/User training
  5. TCO

I too have used a variety of systems both Open Source and Commercial.  Most have good points and bad points and can be reasonably compared.  These 5 steps have been the most important to me.

Level 7

1. Stability of Solution

2. Implementation

3. Reporting - if you can't export relevant data, what good is it...

4. Ease of Use

5. Access to Good Vendor Support (and we all know what "Good" means)

6. Cost

Level 17

I would use the same guidelines as well. It needs to be powerful, easy to use and install.

Level 11

So it seems to me that I have a lot in common with my peers... Stability and ease of implementation are very important among some other things like reporting and cost.

From all the products I've used, SolarWinds has definitely been the easiest to implement and the most stable if you feed and take care of it properly. What have your experiences been?

Level 11

I completely agree with cfwalker8 calliser and jkump

All of these are good indication of a good NMS.

Level 10

calliser I agree with this approach

Level 15

Overall having spent years working with *Nix-based monitoring systems, SW was easy to implement.  Strangely though, most of my peers have only one complaint about SW.  Speed in that the UI is very slow.  Since, I have a bunch of experience with PERL, and PHP web development and how fast those languages make the UI work.  It seems that .Net tends to be slow and the development that I personally done in .Net also seems to be slow.  Beyond that, I find SW to be both technically and economically viable.  The reports work well for my departments.  Was not a huge learning curve to get our help desk staff on board with being able to review the basics before escalating issues.

Level 14

There are many good suggestions there.

However - since using Solarwinds products I would extend those to include :-

  • How are product enhancements decided?
  • How closely do you work with your customers to change you product?
  • How often do new version arrive?
  • How engaged are your customers?
  • Do you have a community of customers?

All of the other points raised only show the state of the application at the time you make your choice.

My questions will give you an indication of what will happen during the lifetime of the product.

Level 11

Very good points mcam. The application could be awesome day 1 when you make the initial purchase but if the company or developers neglect it or don't support the product well enough, it'll only be a matter of time before you're looking for another NMS solution.

I think the idea of thwack is awesome. I love how they keep their community engaged, involved, and most importantly, abreast of news and updates relating to all of SW's products.



Level 14

The number 1 thing I look for is can it do most of what I expect it to do.  Then bang for the buck.  Then expandibility.

Level 12

- Cost : I've seen products with a price tag attached to every little bits and parts, and intricate dependencies that made the purchase of a single module impossible, thus $$$.

- Ease of use : I want to save time managing my network, not spend more time. If it's easier to fire up Excel and copy/paste data than it is to get a graph out of the NMS, what good is it ?

- Did I say cost ?

- Stability : the thing should not fall apart because I asked a report on a ridiculously large data set

- Support : We all know what this is about so I won't comment on that one.

- Hmmm... Oh yeah ! Cost !

Level 18

Short list includes the following

1) scalability - if you can't grow big easily then it is time to look at something else.

2) ease of use - must be intuitive

3) Support - this goes without saying

4) ability to manage multiple maintenance windows across many nodes for different times on different days.  It needs to apply to applications, business services, nodes, alerts, component monitors, etc.

5) data warehouse - have to be able to have good historical data for trouble shooting, capacity planning, seeing if the changes in scale or network links to sites did what was needed over time.

6) cheap DR/DEV licensing costs...about 20% of production product cost or better free.

7) Cost/ROI

😎 good flexible wizard for unmanaging nodes/component monitors/alerts (individually or in bulk) and to be able to schedule multiple unmanage periods at a time for any node/component monitor/alert.

Level 21

I think I would have to agree with the list that Jfrazier has put together.  Of the things in that list I really need to stress "Support".  Any product is only as good as the support behind it.

Level 15

Having the unique situation where I can see monitoring tools in multiple enterprise environments weekly, I would personally rank the following:

  1. Ease of use for the NOC
  2. High level of detail available to the upper level engineers
  3. Ease of reporting and trending for management and stakeholders
  4. Customization options (this should include open access to the database, the web interface, AND a good API)
    • There is no such thing as an out-of-the-box "right tool" for an organization. There is a best tool. Allowing me to hack, slash, modify, and automate that tool is what makes it become the right tool.

Realistically, we are at a point with monitoring that there is not much that any one tool has over the others as far as the basic technology goes. SNMP/WMI/ICMP don't change much. What really sets things aside are how the different offerings "fit" in an organization. Oftentimes, what makes SolarWinds so darn great is that it can either be a simple PnP solution that is stood up and operational in a few hours; or it can be a lifetime career with unending possibilities for the engineers and the business.

Level 14

1. Amount of training required by staff to utilize

2. Reliability

3. Cost

Level 9

I would say first is reliability of the solution, then followed by tools that allows you to accomplish the goals of your organizations NMS, whether it be basic monitoring of up/down, interface utilization, etc.  Finally being able to get it setup how you like and letting it do it's things, have a solution that doesn't require a lot of maintenance/management after the setup is key to me since this isn't my primary/only responsibility.

Level 16

Keeping it as simple and realistic as I can, at least in my world, this is how it plays out for us.

Through the admins/users point of view:

  1. Reliability
  2. Ease of use
  3. Support/Training

Through the people writing the purchase orders:

  1. Money
  2. Does it show us flashing green lights
  3. Do we really need all those things that make it work, or can we just get something we do not need?
  4. Nevermind about the money... If you need it, and it will do 100% what we need, then there is no money in the budget. However, if we do not need it, AND it will not do what we need it to do, then let's get 2 of them.

Okay, so maybe that is not exactly how it goes... but my brain is telling me that is close enough...

Level 8

- Amount of capable data

- Reliability

- Support

- Stability

- Security

- Cost

About the Author
Network Engineer with SCANA