Silicon Triangle
AWS C6 Generation: C6i vs C6a vs C6g
Intel Ice Lake vs AMD Milan vs Graviton2 compute comparison
The workhorse 6th generation introduced Graviton2 to mainstream enterprise workloads. Review the benchmarks and pricing that defined the shift to multi-architecture cloud compute.
1. Architectural & Silicon Specs Head-to-Head
Comparing underlying processor microarchitectures, memory ratios, and Nitro generation features using representative xlarge sizing.
c6ix86_64
C6I Compute Optimized
ProcessorIntel Xeon 8375C (Ice Lake)
Clock Speed3.5 GHz
RAM / vCPU2.0 GiB/vCPU
Max Network50 Gbps
Max EBS IOPS160,000
AWS NitroSupported
Min Rate (c6i.xlarge)$0.1666/hr
Monthly Est.$121.62/mo
Cost per vCPU$0.0416/hr
Cost per GiB RAM$0.0208/hr
CoreMark Benchmark70,311
CoreMark / $422,035
Price Index100% of highest
Compute / $ Value45% of leader
c6ax86_64
C6A Compute Optimized
ProcessorAMD EPYC 7R13 Processor
Clock Speed3.6 GHz
RAM / vCPU2.0 GiB/vCPU
Max Network50 Gbps
Max EBS IOPS240,000
AWS NitroSupported
Min Rate (c6a.xlarge)$0.0935/hr
Monthly Est.$68.25/mo
Cost per vCPU$0.0234/hr
Cost per GiB RAM$0.0117/hr
CoreMark Benchmark86,157
CoreMark / $921,465
Price Index56% of highest
Save 44% vs most expensive
Compute / $ Value99% of leader
c6gARM64 (Graviton)
C6G Compute Optimized
ProcessorAWS Graviton2 Processor
Clock Speed2.5 GHz
RAM / vCPU2.0 GiB/vCPU
Max Network25 Gbps
Max EBS IOPS80,000
AWS NitroSupported
Min Rate (c6g.xlarge)$0.0852/hr
Monthly Est.$62.20/mo
Cost per vCPU$0.0213/hr
Cost per GiB RAM$0.0106/hr
CoreMark Benchmark79,512
CoreMark / $933,239
Price IndexLowest Cost
Save 49% vs most expensive
Compute / $ Value100% (Leader)
2. Side-by-Side Sizing & Pricing Matrix
Direct price and benchmark comparison across matching instance tiers in the lowest-cost region. Lowest hourly cost in each row is highlighted in green.
| Size Tier | vCPU | Memory | c6iLinux Rate | c6aLinux Rate | c6gLinux Rate |
|---|---|---|---|---|---|
| .medium | 1 | 2 GB | — | — | |
| .large | 2 | 4 GB | |||
| .xlarge | 4 | 8 GB | |||
| .2xlarge | 8 | 16 GB | |||
| .4xlarge | 16 | 32 GB | |||
| .8xlarge | 32 | 64 GB | |||
| .12xlarge | 48 | 96 GB | |||
| .16xlarge | 64 | 128 GB | |||
| .24xlarge | 96 | 192 GB | — | ||
| .32xlarge | 128 | 256 GB | — | ||
| .48xlarge | 192 | 384 GB | — | — |
3. Workload Decision Framework
Clear, actionable rules on when each family wins based on workload profile, runtime language, and licensing boundaries.
Recommended:C6g (Graviton2)
When: Production microservices targeting maximum cost efficiency on mature instance types
Why: Widely available in all global AWS regions with proven reliability.
Recommended:C6a (AMD Milan)
When: Existing x86 CI/CD pipelines looking for a ~10% discount over Intel
Why: Drops into existing x86 AMIs without rebuilds at lower hourly rates than C6i.
Recommended:C6i (Intel Ice Lake)
When: x86 workloads with specific Intel compiler flags or legacy drivers
Why: Broadest hypervisor and OS compatibility across older enterprise distributions.
5. Related Architecture Comparisons
Explore adjacent generational comparisons and silicon trade-offs.