Scheduling AWS RDS
Software for scheduling AWS RDS has been around for some time, but – up until recently – using a RDS scheduler has involved a long-winded process that could take an hour or more to stop or start the AWS Relational Database Service. It was still preferable to the copy-delete-create-restore method to stop and start the RDS instances, but not ideal. Now that has all changed.
At the beginning of June 2017, Amazon announced the option of stopping RDS instances and restarting them without losing S3 transaction logs or automatic backups. Although accounts will still be billed for provisioned storage, manual snapshots and automated backup storage while the RDS instances are stopped, the reduction in costs is still significant.
Issues with Writing an RDS Schedule
Although the option of stopping and restarting RDS instances is good news for cost-conscious companies, there is a problem regarding how to stop and restart them. In the latest RDS Guide, instructions are provided on how to execute the process manually through the AWS Management Console, but this relies on developers remembering to stop their RDS instances when they are no longer required and is hardly a reliable way with which to reduce the cost of deploying resources on AWS.
An alternative is to write a script to schedule RDS. Although not difficult, it can be a time-consuming exercise depending on the number of RDS instances deployed on RDS and whether you want them all to stop and restart at the same time. Furthermore, writing an RDS schedule fails to overcome the default action that stopped RDS instances will be restarted after seven days. Writing an RDS schedule may be more reliable than a manual process, but it is not necessarily cost-efficient.
Scheduling AWS RDS with ParkMyCloud
Scheduling AWS RDS with ParkMyCloud is a more reliable and cost-effective method of stopping and restarting RDS instances. Our RDS scheduler function works by first conducting a discovery of all a company´s instances on AWS, and then making recommendations about which ones are suitable for stopping – or “parking” as we call it – based on the names and tags assigned to them.
System administrators can accept the recommendations and set company-wide parking schedules for non-production instances – such as those used for development, testing and staging – or can assign each team its own instances to create individual parking schedules. (In the event a company-wide RDS schedule is applied, it can be “snoozed” to allow access to stopped instances while parked).
The process for creating individual parking schedules is simple and quick. It involves highlighting on a grid the times and days of the week during which the instance(s) should be stopped vs. running. A different RDS schedule can be applied to each individual instance or groups of instances depending on the requirements of each development team. Once the hours have been selected, click “Create”.
The RDS scheduler can also be used to park EC2 instances and auto scaling groups. To make the process simpler still, system managers or team leaders can create a logical group from each constituent part of an application and apply the RDS schedule to the group. Compared to manually stopping or writing an RDS schedule for each process, ParkMyCloud´s solution for scheduling AWS RDS is a no-brainer.
How Much Can I Save with an RDS Scheduler?
Most development teams do not require access to their development, testing and staging instances for more than sixty hours a week. Therefore, assuming you were to park non-production RDS instances outside the hours of 7:00 a.m. to 7:00 p.m. Monday to Friday, you would reduce the cost of database instance hours by almost 65%. However, there are significant indirect savings that can also be made.
The indirect savings come from the single dashboard view that system administrators get over a company´s AWS account. Using this view, administrators can identify underused and unused instances and either move them to a cheaper pricing option or terminate them. There may be further savings to be made by reassigning other instances to cheaper regions or types.
The single dashboard view also enhances accountability. Each RDS schedule can be viewed via the dashboard, allowing system administrators to see how much each RDS schedule is saving and whether it is possible to reduce costs further. This depth of information makes for more accurate planning, forecasting and budgeting. Scheduling AWS RDS has certainly come a long way in a short space of time.
Try Scheduling AWS RDS with ParkMyCloud and See How Much You Can Save
If you are paying for RDS database instance hours you are not using, why not take advantage of a free trial of ParkMyCloud and see how much you can save? Contact us today or click the “Try it Now” button and you could be evaluating our reliable and cost-effective RDS scheduler in your own environment within minutes. There is no obligation on you to continue with our full service once the trial period is over and whatever you save during the trial is yours to keep.
If, at the end of the trial, you are happy with how much you have saved and the control and governance you’ve gained, you have the choice of subscribing to our full service or continuing to use a function-limited version of ParkMyCloud for free. Our full service usually pays for itself within two months, but if you would like further information, speak with our industry-leading customer service team, who will be happy to answer any questions you have about scheduling AWS RDS with ParkMyCloud.