Balanced compute, memory, and network resources for diverse workloads.
High overall performance compute instances for compute-bound packages.
High memory instances for memory-extensive workloads.
High disk throughput instances for storage-intensive workloads.
Instances with powerful GPUs for graphics-intensive workloads and parallel computing.
Instances with field-programmable gate arrays for hardware acceleration and low-latency workloads.
Instances with high I/O throughput for data-intensive workloads.
Physical servers for workloads requiring direct access to the hardware.
Instances with baseline performance and the ability to burst above that for short periods.
Instances powered by AWS Nitro system, designed for burstable workloads and cost efficiency.