Custom Service Broker
Not all services necessarily have to be based on existing solutions. Two of our service brokers were specifically developed with the focus to add value for the operation and the developers to represent.
Great attention was paid to the simple usability as well as the good scalability of the services in order to also meet the requirements of large application landscapes.
“In a world where container and services play an increasing role, it is important to rely on open standards”
Responding to an existing load on services by scaling is a fundamental part of enterprise environments and production environments. Automation of this process is essential to provide smooth, direct and effective scaling. For decision making, metrics from the platforms are collected and interpreted for the resources to be scaled.
service configuration via dashboard and API
This Service Broker offers the possibility to carry out individual configurations. These can be performed either manually or via API.
Multiple Scaling Parameters
The autoscaler allows scaling according to CPU, RAM and number of requests. The combination of these parameters is possible without any problems.
The Autoscaler Service is primarily not bound to a platform, but acts cloud/platform diagnostically. This enables the use across multiple platforms.
The Autoscaler Service is divided into several components and is based on a micro service architecture instead of a monolith. This means that the individual components can be maintained, scaled and modified separately without the entire service stopping.
The Autoscaler service uses easily configurable thresholds to decide when and in which direction to scale. In addition, limits can be set on scaling operations to maintain control over the automatic operations.
In addition to scaling with a fixed step size, the Autoscaler service provides a dynamic step size calculation for applications and resources whose load depends on incoming HTTP requests, in order to minimize the number of scaling steps required. Thus, it is possible to react promptly to occurring loads and to directly dismantle previously installed resources that have been inactive after the load has dropped.
Log Metric Service Broker is a simple and fast solution for applications deployed on Clound Foundry to monitor both log messages of all types and container metrics.
In order to use this service, only the applications to be monitored must be linked to the service, and all logs and metrics are consumed by Cloudfoundrys Loggregator, processed, stored in Elasticsearch and then made available to the user in a specially created dashboard.
The LogMetric Service Broker is multi-tenant, so users can only see the data they are authorized to view.
The service comes along with a front end that allows you to stream log messages live, historicize logs, and search logs. Furthermore, all important container metrics are recorded. Furthermore, the service offers the possibility to create diagrams based on metrics and log data.
The Service Broker supports the authentication model of Cloud Foundry and only allows a user access to the data of an application if he has the corresponding authorizations.
All tools used in connection with the application can be scaled horizontally under increased load to counteract this load.
Service bindings allow you to obtain only logs and metrics of selected applications.
All our service brokers meet the official Open Service Broker API definition. With our interface, developers can order databases in self-service and to a large extent also configure and manage them in self-service (e. g. backup management, specific configuration options for timeouts, users and much more).