A couple of weeks back we have introduced to you our brand new Engine feature and we’ve promised you that it will only get better and better with time.
Today, we would like to share with you a nifty addition to the feature, that we thought might be a useful upgrade for all our customers who know the exact value of their time.
You know how a specific program on the gym treadmill can be selected without clicking on too many buttons to create the needed set up, right? Well, Enable High Availability (HA) is offering you just that - a quick and easy way to apply a set up of multiple components that will increase your Parse Server app's uptime and fault tolerance instantly.
In IT and computing, High Availability refers to maintaining uninterrupted operational status of a system or certain structural elements for a specific period of time. Essentially that means to have your infrastructure designed and set in a way that issues such as system overloading, downtime and single points of failure(SPoF) are handled. This is usually achieved by adding more power and redundancy through scaling vertically and horizontally in a fast and efficient manner.
SashiDo’s interpretation of this term is giving you the option to set your minimum number of engines to 3 with a single simple adjustment or in other words instantly enabling High Availability. This way you’ll increase your fault tolerance through load distribution and removing the risk of having a single engine whose failure could bring down your whole service. To rephrase, you’ll have a spare active server node ready to handle the transferred from the load balancers orphaned traffic in case one or both your other engines go off-line.
We were aiming to make enabling High Availability a piece of cake and we think we did it! A simple toggle button that instantly sets your engine scale’s Min level to 3 was our way to go. This is followed by a prompt to Save & Deploy. Hit the button and a few magical moments after the deployment is completed...there you have it!
In any case, don’t forget that SashiDo is all about flexibility, so if you need more resources you can always play with the Up arrow next to the scale and set it up. The higher you get the safer your app will be!
In the real world, incidents happen and disasters are sometimes inevitable. Unscheduled downtimes are often caused by physical events we have limited control over, such as hardware or software failure or environmental anomalies. Those can have a massive impact on your apps and their performance.
At SashiDo we don’t like crossing fingers and hoping that a storm would not come. You shouldn’t either!
Imagine you have your app running on a solo engine with no horizontal auto-scaling. Then suddenly a huge traffic spike overburdens it and brings it down. All hell breaks loose until the restart - requests can’t be served and your app is also going down…
Now, ask yourself how long your app’s users will wait for a downtime episode to be resolved until they just decide to forget about your services and move on to the next provider? Can they go a day or two tolerating malfunctions or it’s a matter of minutes? For example, back in 2013 Amazon had a downtime episode of 15 minutes(web and mobile) which led them to lose $1 million (about $66 240 per minute) in sales.
This is exactly how important High Availability is and investing in maintaining an uninterrupted user experience is always worth the money!
To wrap things up you should know that all apps on our platform have enabled horizontal auto-scaling by default. We also provide you with an easy to use and optimized tools for implementing redundancy and availability of resources. With this being said know that SashiDo got you covered, so take a breath and check out our new High Availability feature.
And finally, something really important we couldn't miss putting a reminder on - you can’t really have a highly available application if there is no Automated Database Backups enabled. You’ll always need to consider the worst-case scenario and be prepared with a recovery plan. What is better than taking preventative measures in advance and securing your data is available at any given moment?