Managed Services

To guarantee the maximum costs savings possible, we firstly installed our monitoring tool agents to obtain more metrics for the implemented infrastructure in AWS, secondly, we analysed the data that we collected, over a 2 weeks’ time span of deploying the agents.

Additionally, through the monitoring system console, it is possible to create customized or predefined reports, which can be exported to a file in text format, spreadsheet and other more common formats. The data collected by the system can be kept for a period of time to be defined.

The monitoring process is based on metrics and logs collected from the various services, which are collected from the native monitoring systems of each service, or through the installation of monitoring agents.

Magic Beans provided a set of pre-defined reports regarding the use of Logic AWS services based on the monitoring tool used, these were regularly updated, being made available to Logic periodically and/or upon request. Specific reports were implemented according to the client's needs, through their analysis.

Magic Beans makes a report which we call steering where we review all the client’s resources and what can be changed to improve their costs. This report is done on a monthly basis and is the document that summarizes the relevant events and infrastructure status during the previous month, also including an analysis of consumption.

This document is presented at the beginning of the month and it was intended in this way to maintain regular contact with Logic, informing and collecting information in order to promote optimization and plan future actions.

The verification of backup operations is also subject to a monthly report, in order to maintain control over the effective safeguarding of data. This documentation is sent to Logic and is the basis for periodic meetings in which they are discussed, with the intention of analysing the performance of the previous period in this process and proposing optimizations to be implemented.

Following these reports and changes discussed with Logic, the following problems were identified:

• There was a need to re-architecture Logic’s infrastructure, where some EC2 Instances needed to be right sized;
• There was a need to improve plans for backups and it was decided to use AWS Backup, as it lets us automate all these backups in one place by defining one or more backup plans and specifying the instances or group of instances we needed to backup. With AWS Backup we could specify what happens to the backups after several weeks, months or years and we could automatically define an S3 lifecycle policy to move the data to cold storage or delete it when it was no longer needed which was a huge improvement for the backup procedure;
• Also, for Logic to save more in costs each month, we did a reservation of their instances, called Reserved Instances. With this Logic could save up to 52% over equivalent on-demand capacity, Reserved Instances were presented in 3 options – All up-front (AURI), partial up-front (PURI) or no-upfront payments (NURI). Logic decided to go for 1 year reservation no-upfront.