Compute unit (CU) pricing vs. host-based pricing (APM and Infrastructure)

New Relic APM offers a choice between two pricing models: compute unit (CU) based pricing and host-based pricing. New Relic Infrastructure offers only CU-based pricing. This document explains how these pricing models work.

Other related documentation:

Compute unit pricing

Availability of CU-based pricing depends on the New Relic product:

  • New Relic APM: choice of either CU-based pricing or host-based pricing
  • New Relic Infrastructure: only CU-based pricing

With CU-based pricing, your monthly price is determined by the size of your host running New Relic and the number of hours it connects to New Relic during one month. If your host is connected to New Relic at any time during an hour, that hour counts towards the CUs. The host size is calculated by the amount of computing power and memory in the host.

For New Relic APM, CU-based pricing is the best choice if you run a lot of dynamic computing environments in the cloud. For this reason, CU-based pricing is sometimes referred to as cloud pricing.

CUs are calculated as follows:

Accounts_CU-calculation.png

For example:

  • If your host has 2 CPU cores, 2GB RAM, and connects to New Relic for one hour, it consumes 4 CUs.
  • If your host has 2 CPU cores, 2GB RAM, and connects to New Relic for an entire month (equivalent to 750 hours), it consumes 3000 CUs.

You can purchase blocks of CUs to be consumed on a monthly basis. The total number of CUs purchased monthly is calculated by adding up the estimated CU consumption for all hosts for the month. There is no month-to-month rollover of unused CUs. Also, New Relic does not charge by JVMs, containers (such as Docker or Cloud Foundry), or application instances--it charges by the hosts running those containers or application instances.

Price points and CU caps vary, depending on the New Relic product and subscription level. You can view CU-based account usage from the New Relic UI.

Host pricing

Pricing for your New Relic APM account can be either CU-based or host-based. New Relic Infrastructure uses only CU-based pricing.

With host-based pricing, New Relic charges based on the number of equivalent hosts used in a month. One equivalent host is defined as: a host connected to New Relic for 750 hours (the approximate number of hours in a month).

These hours can be divided across more than one host. For example, you might have three hosts that are each connected to New Relic for 250 hours during one month: these hours would add up to equal one equivalent host.

Once connected to New Relic, hosts are distinguished by their unique hostnames. A host is connected to New Relic when the language agent is active and is deployed on the host. New Relic does not charge by containers (such as Docker or Cloud Foundry), JVMs, or application instances--it charges by the hosts running those containers or application instances.

New Relic APM offers you a choice between host-based and CU-based pricing. Host-based pricing is ideal for customers with mainly static environments, consisting of hosts managed by the customer in their own data center.

Host definitions

To understand how New Relic computes both host-based pricing and CU-based pricing, it's important to understand what the word host refers to.

A host can be one of the following:

  • A physical machine is a hardware-based device with dedicated physical resources, including memory, processing, and storage. Each machine has its own OS which applications run on.
  • A virtual machine (VM) is the software implementation of a physical machine that executes programs like a physical machine. One or more virtual machines can run on a physical machine. Each virtual machine has its own OS and allocated virtual machine resources such as RAM and CPU.
  • A cloud instance is a type of virtual machine that is run in the public cloud. In this context, virtual machines and cloud instances are different from Java Virtual Machines (JVMs) and containers.

For more help

Recommendations for learning more: