Advisor

AWS R6g Vs. R5 Instances: A Side-By-Side Comparison

R6g and R5 instances are designed to handle memory-intensive workloads. Although the differences between them are nuanced, here is what sets them apart.

For modern businesses, enterprise applications, such as big data processing in real-time and high-performance databases, are essential. On the other hand, these scenarios can require a lot of computing capacity, the kind that would require substantial cash and a lot of time to build on-premises.

As an alternative, Amazon Web Services (AWS) provides R6g and R5 instances for use in the public cloud. Here’s a five-minute guide to determining which one is right for you.

What Are Amazon EC2 R6g Instances?

The R6g instance family is one of several Memory-Optimized instance types available through the Amazon Web Services (AWS) cloud. With R6g instances, you can process large data sets in memory at high speeds.

The instances use custom, Arm-based, and 64-bit AWS Graviton2 processors. The result is a price-performance ratio up to 40% better than comparable R5 instances when handling memory-intensive workloads.

Amazon EC2 R6g instances also support Enhanced Networking (up to 25 Gbps network bandwidth), Nitro System (dedicated hardware and lightweight hypervisor for optimal cost-performance ratio), fast local NVMe-based Solid State Drives (SSDs), and are EBS-optimized (high-performance block storage on AWS).

What Are Amazon EC2 R5 instances?

R5 instances are an older generation of Memory-Optimized instance types compared to R6g ones. But while R6g instances deliver a maximum of 512 GiB of memory, the largest R5 instance delivers up to 768 GiB for applications requiring lots of memory.

In addition, R5 instances deliver up to 5% more memory per vCPU than previous generation R4 instances. Also, AWS designed R5 instances to deliver up to 10% better price per GiB and 20% higher CPU performance compared to the R4 ones.

How Do R6g And R5 Instances Compare?

	R6g Instances	R5 Instances
Supported processors	Custom-built, Arm-based AWS Graviron2 processors with 64-bit Arm Neoverse core	3.1 GHz (Skylake 8175M or Cascade Lake 8259CL) Intel Xeon Scalable processorsAMD EPYC 7571 for M5a instances
Number of Cores	2 to 64 cores per instance	1 to 48 cores per instance
Maximum Memory	512 GiB	768 GiB
Maximum Storage	EBS and NVMe based SSDs	Up to 64 TiB in a single volume. EBS and NVMe based SSDs
Instance sizes	r6g.medium (1 vCPU and 8 GiB memory)r6g.large (2 vCPU and 16 GiB memory)r6g.xlarge (4 vCPU and 32 GiB memory)r6g.2xlarge (8 vCPU and 64 GiB memory)r6g.4xlarge (16 vCPU and 128 GiB memory)r6g.8xlarge (32 vCPU and 256 GiB memory)r6g.12xlarge (48 vCPU and 384 GiB memory)r6g.16xlarge (64 vCPU and 512 GiB memory)m5.metal (64 vCPU and 512 GiB memory)	r5.large (2 vCPU and 16 GiB memory)r5.xlarge (4 vCPU and 32 GiB memory)r5.2xlarge (8 vCPU and 64 GiB memory)r5.4xlarge (16 vCPU and 128 GiB memory)r5.8xlarge (32 vCPU and 256 GiB memory)r5.12xlarge (48 vCPU and 384 GiB memory)r5.16xlarge (64 vCPU and 512 GiB memory)r5.24xlarge (96 vCPU and 768 GiB memory)R5.metal (96 vCPU and 768 GiB memory)
Networking	Up to 25 Gbps with Enhanced Networking	Up to 25 Gbps with Enhanced Networking
Best Use cases	Real-time big data processing, open-source, databases, and in-memory caches	High-performance databases, real-time big data analytics, mid-size in-memory databases, distributed web scale in-memory caches

When To Use R6g Instances

If you are looking to use one of the newest generation of AWS instances to improve workload performance, while saving up to 40% in costs (compared to using R5 instances), Amazon EC2 R6g instances may be the way to go.

When To Use R5 Instances

Choose R5n instances for use cases that can take advantage of high network throughput and packet rate capabilities, such as distributed web scale in-memory caches.

For high performance databases, R5b instances deliver the fastest block storage performance of all Amazon EC2 instances.

The AMD-based R5a instances are ideal when you want a balance between performance and cost, since they are 10% cheaper than comparable Intel instances.

How To Choose The Best AWS Instance Types For Your Application

Picking the right instance types for your different use cases can be daunting. Doing it manually is even worse; error-prone, time-consuming, and likely to break your budget.

Instead, you can use CloudZero Advisor to rightsize your AWS instances to your workload requirements based on factors such as your budget, region, resource type, and AWS service.

CloudZero Advisor is interactive, intuitive, and handy — check it out here.

Better still, CloudZero is a leader at providing immediately actionable insight into who, what, and why your cloud costs are changing. No one else shows your cost by individual customer, feature, team, project, etc, like we do.

Schedule a demo today to experience CloudZero first-hand.