From the course: AWS Well-Architected Framework: Performance Efficiency Pillar

The performance efficiency pillar - Amazon Web Services (AWS) Tutorial

From the course: AWS Well-Architected Framework: Performance Efficiency Pillar

Start my 1-month free trial

The performance efficiency pillar

- [Mark] Welcome to Performance Efficiency. We're going to be looking at how we tune our performance when we are actually running applications in AWS. But this might not be the best starting point for you if this is the first class you're looking at. There are other classes dealing with each of the pillars, including an overview and you might want to start with the overview. But if you've already looked at all the other movies, thank you and you can continue on with Performance Efficiency. So what are we talking about with performance efficiency? Well, we have to use our computing resources efficiently and computing resources include the compute, the storage, which is a storage array, the database, which is more compute, and the network, which is really compute as well because when you order your EC2 instance, that's when you decide on the speed of the network. So these are all interrelated and you can't have performance efficiency unless you've looked at operational excellence, security, reliability, and the cost optimization. This is really the icing on the cake. And you may find in this class that there is some topics which we've looked at in more detail in the other framework components, the other pillars. So this is looking at, all right, you've got the rest of the work done, now let's get that performance as good as we can. The fine tuning that we have to do. And we want to maintain this efficiency over time. And that's the trick. You might be using something which was great a year ago, even six months ago, but now you have to tune it. So we have to take a look at the changes in our hosted application. And hopefully some of these changes can be automatic such as the changes in demand, meaning that you might want your storage to increase automatically. You might want your compute to increase automatically, say with auto scale. You also want to maintain the efficiency as the overall demands change. They may be longterm. They may be short term. You might have differences in how you use the application. So for example, you might have a very powerful set of instances that you don't need anymore because the application isn't being used as much. You also have to look at what's new at Amazon and adapt new technologies which gives you higher efficiencies. So if you started off with an instance that was great two years ago, undoubtedly there's a better one and a cheaper one available. You just need the time to actually make those changes. So performance efficiency is a long-term project. And as you go through this class, you'll learn how to carry out the steps to achieve your performance efficiency, what you're looking for and maintain it going forward.

Contents