A crowded market…
In the observability market, fierce competition is taking place with the emergence of tools based on open-source projects as well as commercial solutions. Large commercial providers are trying to solidify their dominance in the market by providing not only metric monitoring solutions, but also monitoring services such as log, security, database, APM services, and AI/ML ops.
To compete with these large commercial solution providers, several startups are aiming to differentiate themselves through creative ideas regarding monitoring dashboards and application tracing. Additionally, open source-based service providers are launching managed services that overcome the limitations of OSS projects, including Prometheus’ long-term storage concerns. Examples include the enhanced trace capability offered by Honeycomb.io‘s “BubbleUp” function, the integration of market-leading dashboards such as Kibana and Grafana by Logz.io, as well as the M3-based long-term storage function provided by Chronosphere.
As cloud-native services enter the majority stage commercially, the observability market is also fiercely competitive. So you may wonder – why does NexClipper want to enter this market and what do you bring to the table?
Observability doesn’t just happen. To monitor a resource that requires monitoring, you first need a solution to monitor the observability target and maintain the solution upon any change of the targets. An increase in monitoring targets makes observability more difficult. Even if alert rules are changed, efforts to reflect them may accompany significant resources and business risks due to errors until a separate automation tool is implemented. Additionally, if the target is not Kubernetes-based service, but diversified such as BareMetal and IOT devices, the difficulty of operation further increases.
As you already know, observability is not simply monitoring, but to identify the cause of an incident and apply a solution to solve it. For this purpose, in the past solutions have been continuously adding new monitoring methods and trying to provide more information to users. However, in this flood of dashboards and alarms, it is still difficult to find the root cause of a problem.
To solve this problem, more advanced APM solutions are emerging, and improvement plans, such as applying AI/ML to find the cause of failure, are being developed. Innovation in this field will continue with new services of cloud native technology constantly appearing, and companies that are actively using them increasing.
As these new services are being used commercially, there is also a demand for observability for the services. Therefore, the speed at which observability of a new service can be provided with a verified service is also becoming an important measure for observability. In the case of commercial observability services, it takes time to secure observability of new services. While for open source-based services, you can easily find metric exporters developed by the service community through Github and so on, it is difficult to know how reliable these exporters are until you try and verify them yourself.
NexClipper Observability is an observability service based on open-source projects such as Prometheus, Cortex, Loki, Tempo and Grafana. By automating various tasks from installation to operation, customers can easily utilize the observability service on private and public clouds as well as bare-metals. In addition, metric, log, and trace are integrated to help identify the cause of incidents. We will introduce the features details on this blog in the near future. In this article, find out how NexClipper differentiates itself by providing distinct functions to overcome many of the addressed challenges.
1. Distributed Kubernetes stack management: sudoRy
NexClipper’s solution focuses on how to manage resources in a distributed Kubernetes environment. Automating and standardizing resource management is essential not only for securing observability but also other DevOps operations.
sudoRy, the OSS project of NexClipper, uses the dispatched client at the remote Kubernetes clusters to asynchronously take and execute the service catalog instructed by the sudoRy server, and report the execution results. A service catalog is a standardized template set for provisioning by the Helm chart, executing Kubernetes and Prometheus APIs. This template can be applied directly to the computing zone or can be scheduled, enabling efficient and effective operations. In other words, you can think of it as a service that dispatches workers to distributed data centers, provides standardized manuals, executes them at scheduled times, and reports execution results. For more technical details, please refer to sudoRy, NexClipper’s OSS project on Github.
NexClipper’s observability enables metric-based intelligent operation – these are the main features:
First, it helps you manage who and what channels should receive alerts and notifications. Resources on a per-cluster and per-service basis can be assigned to groups of users to support alerts and analytics only for those resources that are needed.
In addition, alerts and notifications issued from Prometheus and DevOps operations, related metrics, and countermeasures are integrated through an incident to facilitate resolution. The history of incidents will serve the solution of incidents that may occur in the future.
This accumulated information will be the foundation of AI/ML-based operations to be introduced in the future as NexClipper will provide automation (development environment installation, deployment) service for DevOps operation in the future. This history will be a key input for intelligent metric operation.
NexClipper ExporterHub is an OSS that provides Helm charts for exporter installation, as well as alert rules and Grafana dashboard introduction using verified exporters from the community or improved exporters by NexClipper.
NexClipper supports the automatic installation of the corresponding exporter so that users can use it easily through their NexClipper Observability solution. Users can access the review of the exporters, understand the meaning of the metrics and installation steps of the exporters, as well as reduce the burden of creating alerts and dashboards. NexClipper plans to support users to easily use various exporters by upgrading the functions of the current ExporterHub version and providing documentation on detailed functions of each registered exporter.
Hopefully this brief overview provided a good introduction into what NexClipper brings to the table when it comes to cloud native observability and what made us embark on this journey. We would be happy to welcome you onboard for the trip – so feel free to reach out to us if you have any further questions. More updates coming soon!