What is most valuable?
An easy-to-use interface for both admins and users is key. Adoption of any product can be constrained by a UI that is neither intuitive nor visually engaging. The SolarWinds NPM framework, upon which most of their flagship products are built, empowers a wide variety of admins and users to quickly find value in their installed products.
How has it helped my organization?
We used a monitoring platform from an enterprise vendor that was so complex to manage that we struggled even to deliver the base infrastructure monitoring services. This is not a unique position as many IT departments manage by the mantra of "Nobody was ever fired for recommending [insert enterprise vendor here]".
In a few short years, SolarWinds has empowered us to provide robust infrastructure monitoring, including a converged view of our servers, both the physical and virtual servers, and the storage and application environments.
What needs improvement?
- Real-time analytics is the major weakness of the SolarWinds NPM platform today. They've done a fantastic job enabling the collection of data. Users are now demanding the ability to generate their own graphs, views, and so on; all of this must be done by an admin today.
- Scaling to the enterprise level is also an issue. SolarWinds NPM officially supports up to 100K elements (the number is unofficially about 50% higher), but enterprise customers could scale beyond that limit. However, this would only be applicable to environments with extensive interface monitoring requirements (where every interface is an element) or there is a node count > 10,000. SolarWinds is committed to growing that threshold in the near future, but no official announcements have been made.
- Enterprise customers would also like to see high-availability and fault-tolerance. HA v1 was implemented in the latest release at an additional cost. FT has yet to hit the current release but is something we regularly discuss on the Thwack forums.
For how long have I used the solution?
I have been using various versions for the past 10 years.
We are currently using:
- SolarWinds Network Performance Monitor 12.0.1
- SolarWinds Server & Application Manager 6.3.0
- Storage Resource Manager 6.3.0
- Network Traffic Analyzer 4.2.1
- Virtualization Manager 18.104.22.1685
What was my experience with deployment of the solution?
Scalability and HA will impact larger customers. The current HA release requires the redundant nodes to be on the same VLAN. This limitation is being actively addressed in version 12.2 which is in beta testing as of the writing of this review.
Scalability is fit for most customers with capacity for over 10K devices (~100,000 elements). As with all applications running at scale designing your infrastructure is key. The published specs will need to be tuned for your environment at both the polling and database layer to push your implementation beyond 10K devices. (See notes below on scalability)
What do I think about the stability of the solution?
Again, enterprise-scaled customers are probably the only ones to notice issues here and it usually happens post-upgrade. We process 4,000 transactions per second in our monitoring environment and that scale exacerbates any code inefficiencies or bugs. However, the platform is generally very stable and any issues are dealt with rapidly via tier 3 support and the development team.
What do I think about the scalability of the solution?
SolarWinds NPM officially supports up to 100K elements (the number is unofficially about 50% higher), but enterprise customers could scale beyond that limit. However, this would only be applicable to environments with extensive interface monitoring requirements (every interface is an element) or a node count >10,000. SolarWinds is committed to growing that threshold in the near future, but no official announcements have been made.
Virtualization Manager can scale almost infinitely, but we found that it performed best, especially when integrated with the SolarWinds NPM framework, with more moderately sized vCenters rather than a fewer large vCenters.
How are customer service and technical support?
Customer service is exceptional if not a little too communicative at times. SolarWinds does an exceptional job of engaging their customers. Technical Support
Technical support is currently the achilles heel of SolarWinds.
- Non-critical issues can be handled through an extremely active and passionate user forum (at thwack.solarwinds.com), but actual technical support requires engaging their team via a case that can be opened from the support portal or by phone.
- Every customer receives the same level of support regardless of their size, and there is no option to purchase 'advanced' support at this time. Enterprise customers will find this model challenging. This is mitigated by hiring experienced SolarWinds engineers, especially SolarWinds MVPs.
Developing an enterprise support model with measured and reported SLA and OLA targets should be at the top of the SolarWinds support model to do list.
Which solution did I use previously and why did I switch?
In every case where I have used SolarWinds, I have displaced an enterprise platform from a larger software company including, but not limited to, IBM, HPE, NimSoft, and so on. I've also displaced segment competitors because of the aggressive pricing model from SolarWinds and their favorable cost-to-value ratio.
How was the initial setup?
The initial setup is as about as easy as they come. SolarWinds products are renown for being easy to install and configure, and for rapid time-to-value; but the devil is in the details.
As an experienced enterprise engineer, I often see inexperienced engineers jump into the product and start producing metrics that add value under the assumption that everything out-of-the-box adds value. This is not the case; and it is a weakness of the installation model. Customers can be lulled into a false sense of success by their rapid deployment and TTPP (time-to-pretty-pictures). Understanding the data is just as important, perhaps more important, than getting the data.
What about the implementation team?
We implemented through an in-house team but that team consisted of a lead engineer who is now a SolarWinds Head Geek. You can absolutely deploy this without professional services but depending on your timeline, objectives, and workload within your organization you may need to engage the services of a consultant. There are many freelance and corporate freelance options out there but be sure to ask around about their experience.
What was our ROI?
Reduced annual support costs by 75% year-over-year based on our prior solution.
What's my experience with pricing, setup cost, and licensing?
The licensing model is such that you can purchase only what you need; and then grow into the next level by paying only the difference in price and the associated maintenance costs. The first year of maintenance is included in the initial purchase. This helps drive down the ROI in the first year. Subsequent years are a percentage of the retail cost. They are on par with similar service offerings from other companies.
Which other solutions did I evaluate?
We considered options from most major vendors. SolarWinds won out because of the low price-to-value ratio; and the immediate and long-term cost savings made by displacing other products.
What other advice do I have?
Avoid the urge to go quickly. Take the time to forecast your growth in the next 12-24 months and plan the infrastructure for that target. Always follow the size and scaling recommendations, especially for your database environment. Do not use RAID5 for your database as the transactional environment or SolarWinds Orion will overwhelm your storage environment.