When the network traffic for an instance exceeds a maximum, AWS shapes the traffic that exceeds the maximum by queueing and then dropping network packets. You can monitor when traffic exceeds a maximum using the network performance metrics.
EC2 instances such as the C5n are particularly suited to FEA applications due to their balance of CPU, memory and network performance. With 100 Gbps EFA networking, a large number of CPU cores and memory capacity of up to 192GB, these instances provide all the elements necessary to run demanding FEA workloads.
M3 instances are the newest generation of general-purpose instances, and give you the option of a larger number of virtual CPUs (vCPUs) that provide higher performance. M3 instances are recommended if you are seeking general-purpose instances with demanding CPU requirements.
Bandwidth for a current generation instance with less than 32 vCPUs is limited to 5 Gbps. Bandwidth for single-flow (5-tuple) traffic is limited to 5 Gbps when instances are not in the same cluster placement group.
For t2/m3/c3/c4/r3/i2/d2 instances:
m1 small, medium, and large instances tend to perform higher than expected. c1.medium is another freak, at 800 MBit/s.
I gathered this by combing dozens of sources doing benchmarks (primarily using iPerf & TCP connections). Credit to CloudHarmony & flux7 in particular for many of the benchmarks (note that those two links go to google searches showing the numerous individual benchmarks).
The large instance size has the most variation reported:
Burstable (T2) instances appear to exhibit burstable networking performance too:
The CloudHarmony iperf benchmarks show initial transfers start at 1 GBit/s and then gradually drop to the sustained levels above after a few minutes. PDF links to reports below:
t2.small (PDF)
Note that these are within the same region - if you're transferring across regions, real performance may be much slower. Even for the larger instances, I'm seeing numbers of a few hundred MBit/s.
FWIW CloudFront supports streaming as well. Might be better than plain streaming from instances.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With