LogicMonitor Edwin AI

What do you like best about the product?

Senior advised me just to watch the alerts from the server. I'm Data Analyst intern, I don't care about pings on the server. I looked through the metric graphs, and noticed the “Predictive Forecasting” settings. The "Training Window" slider (30 days, 90 days, or customized offset) can be adjusted manually to train this Holt-Winters math model. This is possible even for changing confidence interval! If you're not one of the happy normal users, but you're a math nerd, you can see the HTML5 canvas training forecast line change when you change the training offset (if you don't, you can lock it), and that's pure heaven.

What do you dislike about the product?

This is a beautiful feature that they don't publicise! The training slider is deep within the two sub-menus in widget settings. Other than that, if I export to CSV, it only includes historical raw data and not forecasted coordinate points computed. I don't like having to copy coordinates for my Python script!

What problems is the product solving and how is that benefiting you?

Manager wants to know what will be the first time that our database disk is going to be full? Rather than manually regressing the data for a few hours, I employed this forecast tool. I set training window to disregard the temporary training data spikes last month, and BOOM! It gave exact date (October 14th). I've been blown away by my prediction slide deck. Made my internship by getting a niche feature!

What do you like best about the product?

I'm DevOps engineer, I am measuring everything in numbers. We have been using LogicMonitor for a total of 14 months. Best part is having an agentless collector architecture. We installed collector on 1 standard VM and in 2 hours it discovered more than 650 network devices & servers. This reduces manual set up time by 80% versus previously used agent based tool.
Ring logic is good. We used switches for dynamic thresholds for CPU and memory instead of static thresholds. This was the reason alert noise is going down by 62%. We were receiving 400 alerts a week from our on call team and now we are receiving 150 actionable alerts a week! Full support for integration with Terraform. Monitoring is "As Code". Very low API response time - always <50ms for metric queries.

What do you dislike about the product?

pricing model is strictly on the number of devices. We have both Kubernetes and AWS auto scaling groups in place. Occasionally short-lived containers and or EC2 instances are being started for 1 hour, as a result that is consuming licence quota. Due to this, we are paying 12% more than we expected on our monthly bills.

What problems is the product solving and how is that benefiting you?

Main problem it is solving is hybrid infrastructure visibility. We have 30% on-premise data center and 70% AWS cloud. Both now in a single pane of glass dashboard.
Benefit can only be quantitative:
Now, with the visibility into the exact root cause node in dashboard, Mean Time to Resolve (MTTR) is improved to 26 minutes, down from 45 minutes (42% improvement).
The uptime of the infrastructure is improved from 99.91% to 99.99% for last two quarters.
Capacity planning report made it very easy to identify and downsize 40 idle VMs. This optimization is soaring the savings in cloud billing by a sum of exactly $3,200 a month.
Complete overall ROI is realized after just 8.5 months.

What do you like best about the product?

As a Cloud Engineer, I am currently responsible to manage large multi-cloud (AWS & Azure) infrastructure for my organization. Best thing I like about LogicMonitor is the collector setup. We simply deploy one collector in our VPC and it automatically discovers all cloud resources. Does not require deploying agent on all servers or databases. Works with all cloud services such as ACKS, EKS, RDS and other cloud services.
Dashboards are very interactive and easy to build. I built a lot of custom dashboards for app teams and database teams to view their own metrics. Also, alerting is very fast and reliable. When any one of my pods is crashing or when the CPU hits 90, I receive Slack and PagerDuty messages immediately. Easy to set, but once it is in place it will be totally hands-free.

What do you dislike about the product?

If your cloud environment is extremely large and you have thousands of resources then pricing is a bit on the high side. But sometimes we are getting alarm storm if we don't tune the default thresholds correctly during starting up.
Another thing is writing custom datasources. If default monitoring is not there, we need to write scripts in Groovy for custom datapoints. For beginners, learning Groovy for monitoring is very difficult and slow. Further, if you place too many graph widgets in a single dashboard, UI will be slow loading sometimes.

What problems is the product solving and how is that benefiting you?

We currently check AWS CloudWatch, Azure Monitor, standalone network tool for checking issues. It was big headache to jump between portals. At the moment we have one pane of glass for all.
Helping me find the root cause of an issue very quickly. Application speed is slow—server metrics, network traffic and database load all can be viewed in one place, simultaneously. It saves our lot of time and MTTR (Mean Time to Resolve) is reduced nicely. Obtaining bugs before it's being reported to us.

What do you like best about the product?

We have been using LogicMonitor in my company for some 1 year. The best thing I like is that it can monitor servers, network, cloud (AWS, Azure) and everything in one dashboard. Deploying collectors, and they auto-disonnect everything was really easy. The alerts are extremely speedy and PagerDuty and Slack integration seems to be doing wonder for our on call team. Dashboards are utterly flexible, I created separate dashboards for management, and for technical team. Saves me tons of time locating problems.

What do you dislike about the product?

The loading of the UI with a large number of widgets on the dashboard is sometimes a bit slow. Also, the generation of custom (complex) datapoints/custome scripts for monitoring has some learning curve and in the beginning you might find it not so easy to understand. Also the price is a little bit on the expensive side when have a whole slew of devices to track. Occasionally, thresholds are not well tuned from the beginning, which can cause us false alerts.

What problems is the product solving and how is that benefiting you?

Prior to this, we had 3-4 different tools which we were using for various purposes. We now have a place in central monitoring for all of our monitoring. It provides us excellent information about infrastructure performance. We receive the alert if the server's CPU goes high, before customers complain that their site is down. I use it to get incident resolution done a lot quicker as my team can view metrics and logs all in a single place. It simplifies overall the management of the infrastructure for DevOps teams.

What do you like best about the product?

The one thing that beats LogicMonitor is their agent-less collector architecture. In our organization we have On-Prem servers, Cisco network devices and AWS/Azure in cloud. It's a hassle to install agents on all these VM. They simply placed their collector on a local machine and boom, all of the infrastructure in our network was automatically detected within minutes of LogicMonitor being installed!
The magic part of the autodiscovery is just that. It is leveraging the following without manual configuration: identification of the OS in use, identification of the listening ports and display of CPU, memory and disk metrics in real-time. Also, the pre-built dashboards are highly useful. Our team isn't forced to spend weeks setting up templates. As a manager I really like and use the single pane of glass view to display our SLA status to the management.

What do you dislike about the product?

The cost is for sure the first thing. LogicMonitor is neither affordable nor inexpensive. These are per resource charges and if you are a large environment case of lots of little VMs or lots of Network Switches the cost can quickly escalate. At times it's like we're paying for monitoring plus more!
The other problem is the beep in the beginning. Your inbox will be inundated with thousands of emails and Slack notifications, if you don't take the time to adjust the thresholds. Alert fatigue is very much a concern here.

What problems is the product solving and how is that benefiting you?

Prior to LogicMonitor, our IT ops people were in a all-hands state of chaos. There were various different open-source tools such as Nagios for the servers, another for pinging across network and AWS cloudwatch for cloud. At the time of any outage, my team was spending working hours figuring out, who actually had a problem and at which point.At the time of any outage, my team was working hours trying to ascertain who had the actual problems and where.
No longer is anyone forced to go from one station to another. Correlating a database with network latency or storage IOPs can be done almost instantly with a slow database in the same dashboard. Our MTTR (Mean Time to Resolution) has come down by almost 40%. It has helped our team to avoid support panic on the weekends and helped to make our overall infrastructure much more stable.