Amazon recently announced that businesses could reduce RDS spend by stopping and restarting instances deployed on its AWS Relational Database Service without losing transaction logs or automatic backups. Initially this was received as good news by cost-conscious businesses, who previously had to contend with the tedious copy-delete-create-restore process to stop and restart RDS instances.
However, once the process for stopping and restarting RDS instances through the AWS Management Console was announced, it became clear that the best way to lower RDS spend was to stop and restart each instance manually. Furthermore, if an instance was stopped for more than seven days, the instance would be restarted automatically and another manual process would be required to stop it again.
In terms of how to reduce RDS spend, this process is not ideal for businesses seeking long-term savings. It relies on developers remembering to stop their non-production instances when they are not required, and remembering which instances they may have stopped seven days previously. Within days of Amazon´s announcement, developers were busy writing scheduling scripts.
Why Scheduling Scripts May Not Lower RDS Spend as Much as You Think
Writing scheduling scripts, maintaining them and revising them as necessary may satisfy businesses with a “we can build, so why buy” mentality, but it is not always the most cost-effective approach to lower RDS spend. Indeed, once you take into account the developers´ salaries and delays in the delivery of core projects, writing, maintaining and revising scheduling scripts could cost more than businesses save.
This is more than likely the case if developers have to write, maintain and revise scheduling scripts for instances that need stopping and restarting at irregular intervals, or if access to the instances is required outside of normal working hours. For this reason, we have developed a product – ParkMyCloud – that can quickly and easily set interruptible parking schedules in order to lower RDS spend by up to 65%.
Reduce RDS Spend Cost-Effectively with ParkMyCloud
In order to reduce RDS spend with ParkMyCloud, account managers create an IAM role for our lightweight SaaS platform so it can conduct a discovery of all the business´s RDS instances. Once the discovery is complete, ParkMyCloud makes recommendations about which instances are suitable for scheduling (or “parking”) based on the names and tags assigned to them.
Account managers first have the options of accepting ParkMyCloud´s recommendations, selecting specific groups of instances for parking, or selecting individual instances for parking. Then they have the options of parking the recommended/selected instances using one of the preconfigured parking schedules provided, or creating a unique parking schedule to meet their requirements.
Different parking schedules can be assigned to different groups of instances – or individual instances – with the click of a mouse; and, if access to an instance is needed while stopped, the schedule can be interrupted (or “snoozed”) with another click of a mouse. Safeguards exist to automatically restart the parking schedule, to prevent the event that developers forget once access to the instance is no longer required.
Based on a working week of 8:00 a.m. to 8:00 p.m., Monday to Friday, businesses that park their non-production RDS instances with ParkMyCloud outside of these hours will save almost 65% of the cost of database instance hours. However, if a permanent “always parked” schedule is selected, and the schedule snoozed when access to the parked instances is required, the savings would be much more.
Additional Ways in Which ParkMyCloud Helps Reduce RDS Spend
At this point we should mention that Amazon´s 7-day automatic restart process does not work on parked RDS instances. When ParkMyCloud parks an instance, it stays parked. Furthermore, there are additional ways in which ParkMyCloud can reduce RDS spend due to the single dashboard view an account manager gets after ParkMyCloud has conducted its discovery of the business´s accounts.
ParkMyCloud´s UI displays not only the RDS instances suitable for parking, but AWS EC2 and auto-scaling groups as well. If the business deploys instances and VMs on Microsoft Azure and Google Cloud Platform, these are also visible. Account managers can use this information in order to create logical groups and park constituent parts of an application simultaneously – and further reduce RDS spend.
The single dashboard view also improves IT governance and helps identify unused instances and VMs which can then be terminated. Account managers can see what parking schedules have been set for each instance, the lower RDS spend created by each parking schedule, and whether the parking schedules should be changed to reduce RDS spend further.
See by How Much You Can Lower RDS Spend in Fourteen Days
If you would like to reduce RDS spend by up to 65%, why not try a free trial of ParkMyCloud and see by how much you can lower RDS spend in fourteen days? Our free trial is intended to give businesses the opportunity to test the merits of ParkMyCloud in their own environments and become accustomed to the schedule process and the improved IT governance ParkMyCloud provides.
To take advantage of this opportunity click the “Try it Free” button on this page or contact us with any questions you have. At the end of the trial period, you have the options of subscribing to one of our competitive payment plans, continuing to use a feature-limited version of ParkMyCloud for free, or forgetting all about us and going back to paying Amazon for database hours you are not using.