c6a.2xlarge

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

If you have a compute-intensive application — maybe scientific modelling, intensive machine learning, or multiplayer gaming — these instances are a good choice.

I am running a network-intensive task that sends several thousand pings/traceroute packets per second to external IP addresses. I've noticed that the network throughput is high soon after creating the instance, but drops off exponentially after the ping process has been running for more than an hour or so. I've tried this on a variety of instance sizes. Even on a c6.2xlarge (whose stated baseline network bandwidth is 2.5 Gbps), it drops down to around 200 packets a minute. Any ideas why this might be happening? Is there any way I can reserve more bandwidth for my instance?

Ah, I'm having the same problem! Which C series did you pick?

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.

I've been using c6i instances in production for a couple months now, in the us-east-1 region. I replaced c5 instances. I'm currently looking at PHP performance on the c6a and we might switch to those if they perform comparably to the c6i. The c5a instances are not great for applications that are sensitive to memory latency, but do work well with more throughput oriented things like Kafka. Edit: looks like c6i is going to be the better option over c6a.