EC2 Instance Types and Families

Choosing the right instance type is one of the most important decisions you make when working with EC2. AWS offers over 750 instance types across multiple families, each optimised for a different workload profile. This lesson breaks down the naming convention, the major families, and how to select the right type for your use case.

---

Instance Type Naming Convention

Every EC2 instance type follows a structured naming pattern:

<family><generation>.<size>

For example, m7g.xlarge:

Component	Value	Meaning
Family	m	General Purpose
Generation	7	Seventh generation
Processor	g	AWS Graviton (ARM-based)
Size	xlarge	4 vCPUs, 16 GiB RAM

Common processor suffixes:

Suffix	Processor
(none)	Intel (default)
a	AMD EPYC
g	AWS Graviton (ARM)
i	Intel with higher-frequency cores

Common additional attribute suffixes:

Suffix	Meaning
d	Local NVMe instance store included
n	Enhanced networking
e	Extra memory
z	High-frequency cores
flex	Flexible (choose vCPU and memory independently)

---

Instance Sizes

Within each family and generation, instances come in multiple sizes. Each step up roughly doubles the resources:

Size	Typical vCPUs	Typical RAM
nano	1	0.5 GiB
micro	1	1 GiB
small	1	2 GiB
medium	1–2	4 GiB
large	2	8 GiB
xlarge	4	16 GiB
2xlarge	8	32 GiB
4xlarge	16	64 GiB
8xlarge	32	128 GiB
12xlarge	48	192 GiB
16xlarge	64	256 GiB
24xlarge	96	384 GiB
metal	Full host	Full host

The metal size gives you the entire physical server with no hypervisor overhead — useful for workloads that need bare-metal performance or hardware-level features.

---

Instance Families

General Purpose (M, T, Mac)

General Purpose instances provide a balanced mix of compute, memory, and networking. They are the best starting point for most workloads.

Type	Use Cases
M7g / M7i / M7a	Web servers, application servers, backend services, small databases, dev/test environments
T3 / T3a / T2	Burstable workloads — development, testing, micro-services with variable load
Mac1 / Mac2	macOS workloads — iOS/macOS app builds, testing

Burstable instances (T family) earn CPU credits when idle and spend them when they need to burst above the baseline. This makes them very cost-effective for workloads that are mostly idle but occasionally need full CPU.

T3 Size	Baseline CPU (%)	CPU Credits / Hour
t3.nano	5%	6
t3.micro	10%	12
t3.small	20%	24
t3.medium	20%	24
t3.large	30%	36
t3.xlarge	40%	96

Tip: Enable Unlimited Mode on T instances to burst beyond your credit balance (you pay for the additional usage per vCPU-hour).

Compute Optimised (C)