Overall instance bandwidth quotas

As a general guide, the smaller sizes of C5, M5, and R5 instance types can sustain up to 10- Gbps network performance. Larger instance sizes can sustain between 10–25 Gbps. Smaller sizes of C5N provide up to 25 Gbps with the largest C5n instances scaling up to 100 Gbps. Some examples of instance type, configuration, and network performance:

Table 1 – c5 and c5n instance family configuration and performance comparison

Model	vCPU	Mem (GiB)	Network Performance (Gbps)	Model	vCPU	Mem (GiB)	Network Performance (Gbps)
c7g.large	2	4	Up to 12.5	c6n.large	2	4	Up to 25
c7g.xlarge	4	8	Up to 12.5	c6n.xlarge	4	8	Up to 25
c7g.2xlarge	8	16	Up to 15	c6n.2xlarge	8	16	Up to 25
c7g.4xlarge	16	32	Up to 15	c6n.4xlarge	16	32	25
c7g.8xlarge	32	64	15	c6n.8xlarge	36	72	50
c7g.12xlarge	48	96	22.5	c6n.12xlarge	48	96	75
c7g.16xlarge	64	128	30	c6n.16xlarge	64	128	100

Aggregate bandwidth throughput for instances between Availability Zones (within a VPC) or between instances in a peered VPC scenario is 25–100 Gbps, depending on instance type (see Table 1). Similarly, aggregate bandwidth to VPC endpoints, such as Amazon S3, is 25- 100 Gbps. Single TCP flow is limited to 10 Gbps for instances in the same placement group and 5 Gbps between instances anywhere else. (TCP flow is defined as traffic going through a single TCP port.) A placement group is a logical grouping, or cluster, of instances within a single Availability Zone, that allows applications to use low latency 10-Gbps network. For more information, see Amazon EC2 instance types.