AWS, Cloud Computing

4 Mins Read

Best Practices for Optimizing Amazon EC2 Spot Instances usage – Part 2


AWS Spot Instances are economical for running elastic workloads on Amazon Web Services (AWS). 

In the previous part of “Best Practices for Optimizing Amazon EC2 Spot Instances usage – Part 1”, we discussed the two key features of optimizing the Amazon EC2 Spot Instances. In this second part, we will learn the remaining two key features.


The instances are not necessarily Amazon EC2 instances but unused instances offered at a discounted rate. They are an effective way to reduce the cost of compute resources, particularly for workloads that can be paused or are not necessarily time-critical.

Before starting with the next two features, let’s discuss ‘Why we should use Amazon EC2 Spot Instances?’

  • Low, predictable prices.
  • Massive Scale possibility.
  • Easy to use.

Pioneers in Cloud Consulting & Migration Services

  • Reduced infrastructural costs
  • Accelerated application deployment
Get Started


In the previous blog, we discussed two Spot Instances best practices that can help us optimize your Spot Instance usage to maximize savings: ‘Diversifying your instances’ & ‘Considering the attribute-based instance types of selection’. This part will discuss ‘Leverage Spot placement’ & ‘Use of the price-capacity-optimized allocation strategy’.

Best Practices

  1. Leverage Spot placement scores:
    Now that we have stressed the importance of flexibility, selecting instances, and finding the best time and location to launch the Spot Instances, let’s look at how to find the preferred time and location. Spot Instances are Amazon EC2 instances with unused capacity. Because of this, the capacity of the Spot Instances can fluctuate. Therefore, we may not always be able to acquire the specified capacity at the specified time that we require through the use of Spot Instances. The Spot placement scores are an element of Spot Instances, which indicates the likelihood of obtaining the required Spot capacity in a particular Region or Availability Zone. Using the Spot placement score, we can reduce the number of times a Spot Instance is interrupted, increase the capacity, and determine the best configurations for running workloads on a Spot Instance. However, it’s important to remember that Spot placement scores are just suggestions and can change depending on your current capacity. They don’t guarantee you will have enough capacity or that you won’t be cut off.Spot placement scores are typically incorporated into deployment automation as a near-real-time tool. However, its logging and graphical capabilities may be useful even before a workload is deployed to the cloud. If you want to know how often your workload has received a spot placement score, you should look at the spot placement score tracker. It’s a tool that makes capturing spot placement scores easy and storing them in Amazon CloudWatch. You can get the tracker from AWS Labs, a GitHub tool repository.When considering the optimal time to deploy Spot Instances and exploring various options via the Spot placement score, it is important to ensure that Spot Instances are launched during peak hours – or when there is a lower level of Amazon EC2 Instance demand. As you may have guessed, there is less idle capacity – known as Spot Instances – during normal business hours than after business hours. Therefore, to maximize the utilization of the available Spot capacity, you should consider running your workload at times when there is less demand for Amazon EC2 Instances and thus, more availability for Spot Instances. So, for example, you might want to run your SPIs in off-peak regions – or regions that aren’t open for business at that particular time.

    In addition, to optimize the utilization of the Spot Instances, it is recommended to utilize previous generation instances if they meet the workload requirements. This is because, similarly to off-peak versus peak hours, previous generation instances typically have more capacity than current generation instances because most users use current generation instances for compute requirements.

  2. Use the price-capacity-optimized allocation strategy:
    Once you’ve chosen a bunch of different and adaptable cases, you need to decide how you will allocate them. When we launch instances, our Autoscaling group will use the allocation strategy we have set to select the specific Spot pool from all our available pools. Spot provides four allocation strategies: Price-Capacity-Optimized, Capacity-optimized, Capacity-Optimized-Prioritized, and Lowest-Price. Each strategy selects Spot Instances within pools based on the price, capacity, prioritized list, or a combination thereof.The Price-Capacity-Optimized strategy was initiated in November 2022 and allocates SPIs based on the highest capacity available at the least cost. It enables Auto Scaling groups to locate the Spot pools that have the most capacity available for the volume of instances being launched. Essentially, if we choose this allocation methodology, we will locate the Spot capacity pools we consider to have the least likelihood of downtime shortly. Our Auto Scaling groups then request Spot Instances from the lowest priced of these pools.It is suggested that the majority of workloads that are executed on the Spot Instances should be leveraged through the use of a price-capacity-optimized Allocation Strategy. To gain insight into how the Price-Capacity Optimized Allocation Strategy selects the Spot Instances, compare it to the lowest-price and capacity-optimized Allocation Strategies.


If you’ve gone through the various Spot Instances workshops we discussed in this blog post and created resources, clear them out so you don’t have to pay for them in the future.


In this blog, we talked about four Spot Instance best practices to help you get the most out of your instances and save money: diversify your instances, think about picking instances based on attributes, use Spot placement scores to get more out of your Spot Instances, and use the Price-Capacity-Optimize allocation strategy.

Drop a query if you have any questions regarding Amazon EC2 and we will get back to you quickly.

Making IT Networks Enterprise-ready – Cloud Management Services

  • Accelerated cloud migration
  • End-to-end view of the cloud environment
Get Started

About CloudThat

CloudThat is an official AWS (Amazon Web Services) Advanced Consulting Partner and Training partner, AWS Migration Partner, AWS Data and Analytics Partner, AWS DevOps Competency Partner, Amazon QuickSight Service Delivery Partner, AWS EKS Service Delivery Partner, and Microsoft Gold Partner, helping people develop knowledge of the cloud and help their businesses aim for higher goals using best-in-industry cloud computing practices and expertise. We are on a mission to build a robust cloud computing ecosystem by disseminating knowledge on technological intricacies within the cloud space. Our blogs, webinars, case studies, and white papers enable all the stakeholders in the cloud computing sphere.

To get started, go through our Consultancy page and Managed Services PackageCloudThat’s offerings.


1. How can I calculate my new vCPU limit?

ANS: – You can find the vCPU mapping for each Amazon EC2 Instance Types to compute the total vCPU limit requirements for your Amazon Web Services account.

2. How can I view my current Spot Instance limits?

ANS: – You can find your current Spot Instance limits on the Amazon EC2 Service Limits page in the Amazon EC2 console.

WRITTEN BY Guru Bhajan Singh

Guru Bhajan Singh is currently working as a Software Engineer - PHP at CloudThat and has 6+ years of experience in PHP. He holds a Master's degree in Computer Applications and enjoys coding, problem-solving, learning new things, and writing technical blogs.



    Click to Comment

Get The Most Out Of Us

Our support doesn't end here. We have monthly newsletters, study guides, practice questions, and more to assist you in upgrading your cloud career. Subscribe to get them all!