EC2 Network Performance of t3

Andreas Wittig – 12 Oct 2018

EC2 Network Performance t3

AWS introduced the t3 instance type in August. Time to update the EC2 Network Performance Cheat Sheet. The following table shows the network capabilities of t3 instances. Compared to the last generation t2 there is no difference in the baseline throughput. However, the burst throughput is more than 5 times higher compared to t2 instance types.

INSTANCE TYPE Baseline (Gbit/s) Burst (Gbit/s)
t3.nano 0.03 5.06
t3.micro 0.06 5.09
t3.small 0.13 5.11
t3.medium 0.25 4.98
t3.large 0.51 5.11
t3.xlarge 1.02 5.11
t3.2xlarge 2.04 5.11

Looking for other instance types? See EC2 Network Performance Cheat Sheet.


Looking for a new challenge?

  • tecRacer

    Cloud Consultant

    tecRacer • Premier AWS Consulting Partner • Germany, Austria, Spain, and Switzerland
    AWS only Infrastructure as Code EC2 Containers Serverless
  • tecRacer

    Cloud Migration Specialist

    tecRacer • Premier AWS Consulting Partner • Germany, Austria, Spain, and Switzerland
    Lift&Shift Transformation EC2 RDS VPC

Methodology

I have run the network performance benchmark in us-east-1 on September 25th, 2018 using iperf3. Each benchmark run took 60 minutes, and a data point was recorded every minute. A fresh c5.18xlarge was used as the counterparty for the network benchmark of each instance type. I have published the source code of the network benchmark at widdix/ec2-network-benchmark.

Explanation

The data analysis in the table shows the burst and baseline throughput of each instance type. The burst throughput is defined as the 95th percentile, which means the throughput was at least reached for 3 minutes. The baseline throughput is defined as the 10th percentile, which means the throughput was reached at least for 54 minutes. The raw data is available as Google Spreadsheet.

Disclaimer

Please note, that the network benchmark was just a random sample and does not claim to be 100% accurate. Therefore, you should treat the EC2 Network Performance Cheat Sheet as the first guide when choosing an instance type based on your network throughput needs. If network throughput is critical to your infrastructure, you should run network benchmarks yourself to verify my results.

Become a cloudonaut supporter

Andreas Wittig

Andreas Wittig ( Email, Twitter, or LinkedIn )

We launched the cloudonaut blog in 2015. Since then, we have published 345 articles, 45 podcast episodes, and 37 videos. It's all free and means a lot of work in our spare time. We enjoy sharing our AWS knowledge with you.

Please support us

Have you learned something new by reading, listening, or watching our content? With your help, we can spend enough time to keep publishing great content in the future. Learn more

$
Amount must be a multriply of 5. E.g, 5, 10, 15.

Thanks to Alan Leech, Alex DeBrie, ANTHONY RAITI, Jaap-Jan Frans, Jason Yorty, Jeff Finley, Jens Gehring, jhoadley, Johannes Grumböck, John Culkin, Jonas Mellquist, Juraj Martinka, Kamil Oboril, Ken Snyder, Ross Mohan, Ross Mohan, sam onaga, Shawn Tolidano, Thorsten Hoeger, Todd Valentine, and all anonymous supporters for your help! We also want to thank all supporters who purchased a cloudonaut t-shirt.