Cloud scale intelligent monitoring & analytics platform

Oddeye monitoring suite is one of the most advanced, all-in-one monitoring and anomaly detection systems available today. It is provided as Software as a Service (SaaS) platform and has been created with intuitive usability as the main priority while offering feature rich dashboards. Configuring the parameters is an effortless process with full support from the OddEye team and video manuals.

Under certain circumstances, it can be hard or even practically impossible to configure the collection of useful metrics using static thresholds for alerts. The reason for this is mostly due to the nature of particular metrics. How will you take into account the time of the day, peak or off-peak periods?

For instance, consider a scenario in which the users of your service are from different geographical regions. In such a situation system statuses and server loads can be significantly different depending on the time of the day. High CPU load is expected and therefore it is normal during peak hours. However, if the system reports high CPU load during off-peak hours this will be considered an anomaly and vice versa.

Setting the correct thresholds manually can be a daunting task for the kind of scenarios mentioned above. Especially when you need to manage a large infrastructure. OddEye monitoring suite uses state-of-the-art Machine Learning algorithms for anomaly detection and setting of thresholds for alerts. These will be triggered when the value of a particular metric is higher than expected but also in situations where the value is lower than where it should be. Taking the CPU metric again into consideration it would be normal to expect a load of 80% during peak hours. What if the load is only 20%? This may be due to network or DNS related problems among many other reason. OddEye will detect this automatically without human intervention and trigger an alert.

In this image you can see two anomalies indicated with two purple balloons with unusually high and low values during a certain period of time. Both of these will trigger an alert. However, OddEye does not rely on Machine Learning only. It is also possible for the user to configure manual thresholds which will be shown on the dashboard as a “Special Alert”.

A good example to illustrate this can be the System Load Average. Regardless of the time of the day, Load Average of Linux systems in production should be below the core count of the CPU. A Quad-Core CPU with Hyper-Threading enabled will have 8 logical cores. As a consequence, the Load Average should be less than 8. Otherwise, system queue will be overloaded and users of the service will experience a slowdown which in turn will make the queue even larger.

This next image illustrates both static (manual) and Machine Learned alerts. Here, the first row represents alerts which are triggered by manually set thresholds and the second row is fully defined by Machine Learning algorithms.

The OddEye web application as well as the OddEye agent are fully customizable. Depending on your specific infrastructure or situation you can configure both manual and Machine Learned alerts.

Advantages of using OddEye Monitoring Suite

Pay-As-You-Go model. Pay for what you use.

No long-term commitment necessary. Cancel anytime.

Fully configurable metrics collection rate down to two seconds.

Minimum one year data retention.

No automatic down sampling of data.We will never auto-average so you can see your full data at any point in time.

Fully configurable alerting and Machine Learning thresholds with intelligent defaults.

Build great looking dashboards using Line, Funnel, Gauge, Pie & Donut, Table, Heatmap and Single Metric charts.

Advanced dashboard query builder with intuitive design.

Raw JSon query builder for advanced users and cases.

Dynamically changing thresholds using Machine Learning.

Open source agents.

Over sixty integrations. New ones are added on a regular basis.

Open API for building your own agents.

Pre-configured dashboard templates for easy setup. Simply clone one and set several parameters to start in minutes.

Enterprise support and creation of non-standard checks for your specific requirements using advanced raw Json Query builder.

Sign-up today and we will add $50 credit to you balance which you can spend to monitor your servers.