AWS recently announced the 7th generation of their Compute instance family. These are the first instances powered by the new AWS Graviton3 processors.

The Graviton processors offer better performance at a lower cost and additional capabilities.

In this article, we will compare the C7g instance family with the Graviton3 processors to the C6g instance family with the previous generation of Graviton2 processors.

Comparing the C7g and C6g instance families

Both C6g and C7g instance families are optimized for compute-bound applications.

The key difference between the two instance families is that the C6g instances are powered by Arm-based AWS Graviton 2 processors whereas C7g instances are based on the newer Graviton 3 processors.


We compared the performance between these instance families by running a bunch of tests on the following instances:

  • c6g.xlarge
  • c7g.xlarge
Instance Type vCPUs Memory (GB)
c6g.xlarge 4 8
c7g.xlarge 4 8

We will be using Sysbench to run various tests to measure the CPU, and Memory performance.


We ran the CPU benchmark using the following command: sysbench cpu --cpu-max-prime=20000 --threads=8 run. We will be comparing the total time as well as the total number of events for the benchmark from start to end.

As we can see from the results below, c7g.xlarge performed much better as compared to c6g.xlarge.


CPU speed:
    events per second:  4426.78

General statistics:
    total time:                          10.0014s
    total number of events:              44279

Latency (ms):
         min:                                    0.89
         avg:                                    1.81
         max:                                   20.91
         95th percentile:                       12.98
         sum:                                79932.41


CPU speed:
    events per second:  2181.39

General statistics:
    total time:                          10.0026s
    total number of events:              21822

Latency (ms):
         min:                                    0.91
         avg:                                    3.66
         max:                                   48.92
         95th percentile:                       12.98
         sum:                                79890.47


We ran the Memory benchmark using the following command: sysbench memory --memory-block-size=1M --memory-total-size=100G --num-threads=8 run. We will be comparing the Operations performed for the different instance types.

In this benchmark, c7g.xlarge performed much better than c6g.xlarge. The C7g instances are the first to feature DDR5 memory which is much faster than the DDR4 memory used in the C6g instances.


General statistics:
    total time:                          1.2762s
    total number of events:              102400

Latency (ms):
         min:                                    0.05
         avg:                                    0.09
         max:                                   20.05
         95th percentile:                        0.05
         sum:                                 9553.29


General statistics:
    total time:                          4.0255s
    total number of events:              102400

Latency (ms):
         min:                                    0.08
         avg:                                    0.31
         max:                                   32.07
         95th percentile:                        0.08
         sum:                                31861.76


We compared the prices between these instance families by looking at the price per hour for each instance type.

C7g instances are more expensive compared to their C6g counterparts.

Price/core for Linux on-demand instances:

Instance Type Price / hour
c7g.xlarge $0.1450
c6g.xlarge $0.1360


The new C7g instances provide much better performance as compared to the C6g instances. They are more expensive than the C6g instances. However, their overall performance to price ratio is much better than the C6g instances.