AWS r5.4xlargevsAWS r5d.4xlarge

r5.4xlarge pairs 16 vCPUs with 128GB of RAM at $1.0080/hr On-Demand (about $726/mo at 24×7). r5d.4xlarge pairs 16 vCPUs with 128GB at $1.1520/hr (~$829/mo). r5.4xlarge is 14% cheaper per hour than r5d.4xlarge ($0.1440/hr gap).

Both are generation-5 memory-optimized instances, but they run on different silicon: **r5.4xlarge** is Intel Xeon (x86_64), **r5d.4xlarge** is Intel Xeon (x86_64). AMD variants (suffix `a`) are typically 10% cheaper than Intel siblings at comparable single-thread performance. Graviton variants (suffix `g`) are usually 20–40% cheaper but require ARM64-compatible binaries — most modern Linux stacks are fine, but verify any compiled extensions, native modules, or third-party binaries before migrating. Same vCPU/RAM ratio, same network performance class, different processor.

On raw price-per-performance, the two are r5.4xlarge delivers ~18% more single-thread Sysbench score per dollar (1084 vs 916 points per $1/hr). That's the cleanest signal we have for "which one runs your workload faster per dollar," but it only matters if your workload is single-thread-bound; for parallel workloads the multi-core scores (13459 vs 12851) are what to weigh. Spot pricing flips many of these comparisons — when r5.4xlarge drops to $0.2814/hr and r5d.4xlarge drops to $0.5869/hr, the cheap-per-hour winner can swing meaningfully.

In practice, pick r5.4xlarge when your workload is closer to memory-optimized (memory-bound work — in-memory databases, real-time analytics, large caches). Pick r5d.4xlarge when it's closer to memory-optimized (memory-bound work — in-memory databases, real-time analytics, large caches). When neither side is obviously right, the cheaper hourly rate usually wins for fault-tolerant batch workloads, while the higher single-core score usually wins for latency-sensitive web traffic. The regional pricing tables linked from each instance page below show where each is currently cheapest — sometimes a >20% regional gap flips the comparison entirely.