Reserved Instances Overview

Introduction

This feature is supported on AWS EC2 only.

Reserved Instances (RI) is a method of purchasing a cloud reserve to reserve the cloud resource for a specific period. RIs offer the ability to significantly reduce instance costs over a defined term, thus, benefitting from the capacity reservation for predictable usage or workloads.

RIs offer up to 80% discount over on-demand instance costs, depending on the cloud provider, payment terms and duration. RIs is one of the most popular ways for enterprises to get great discounts on computing costs. 

 By combining RIs with on-demand instances, organizations can save on running costs without sacrificing reliability and flexibility.

Terminology

The following table explains the RI terminologies.

TermDefinition
ReservationA commitment made by a customer to the cloud service provider (AWS, in this case) for using resources for a defined period. The cloud service provider in honor of this commitment offers a discount to the customer. Depending on the duration and payment terms, discounts may vary from 35% to 80%.

Utilization

The actual duration (in hours and percentage) that a RI subscription was used for a selected period. Once a reservation is purchased, you must provision or run instances of a matching type to benefit from the purchased RI hours and applicable discounts. Unused hours are not carried forward or accumulated.

Float

Transfer of utilization from one account to another. For example, if you reserve instance in one account and the instance is not used for an hour in that account, the usage can be applied to another account.

To take advantage of Float, you must have consolidated billing enabled because Float is limited to the billing account and the linked accounts.

Convertible Reserved Instances

Exchange one or more RIs for another with a different configuration. There is no limit on the number of times you can exchange a RI, as long as the target Convertible Reserved Instance is of an equal or higher value than the Convertible Reserved Instances that you are exchanging.

Payment Methods

Reserved Instances can be purchased for a period of 1 year to 3 years for these resources:

  • Instance Type

  • Region

  • Duration

  • Payment Terms

  • Operating System

  • Availability Zone (Optional)

For each resource, the payment methods are as follows:

  • Full Upfront – The entire amount for the duration of the RI term is paid in advance, providing you with a large discount. There are no per hour charges.

  • Partial Upfront – A low amount is paid to reserve the instance. A discounted hourly rate is applied for the duration of the instance.

  • No Upfront – No upfront payment. A discounted hourly rate is applied during the duration of the instance term.

For partial or no upfront payment, the hourly cost for the RI is applied irrespective of whether the instance is running and the cost of the instance is charged to account as reservation charges.

Reports

Cost Optimizer provides a variety of out-of-the-box reports that help you to track and to manage the performance and status of RI investments in your organization. The RI reports allow you to do the following:

  • Assess ROI on purchased RIs

  • Discover underutilized RIs to optimize usage

  • Uncover opportunities for additional RI purchases for maximum savings

RI helps in answering the following critical questions that you may have about your RIs. 

  • How much did I save from using RIs?

  • How many RIs should I purchase?

  • How can I optimize my RIs?

  • What would be the additional on-demand usage that I can convert to RI to reduce my cost?

  • How are my RIs performing?

  • Which groups in my organization use RIs?

  • Is float applicable to the groups in my organzation that use RIs?

  • What is the utilization of RI across departments? 

  • How much of my instances are running as RIs? 

This is not an exhaustive list, but a compilation of common questions that you may have on RIs.

 The list of RI reports are as follows:



  • No labels
© 2017-2019 Cisco Systems, Inc. All rights reserved