Today, Pinterest’s memcached fleet spans over 5000 EC2 instances across a variety of instance types optimized along compute, memory, and storage dimensions. Collectively, the fleet serves up to ~180 million requests per second and ~220 GB/s of network throughput over a ~460 TB active in-memory and on-disk dataset, partitioned among ~70 distinct clusters.
Most instances are r5 ones:
Instance Size | vCPU | Memory (GiB) | Instance Storage (GB) | Networking Performance (Gbps) | EBS Bandwidth (Mbps)
-- | -- | -- | -- | -- | --
r5.large | 2 | 16 | EBS-Only | Up to 10 | Up to 4,750
r5.xlarge | 4 | 32 | EBS-Only | Up to 10 | Up to 4,750
r5.2xlarge | 8 | 64 | EBS-Only | Up to 10 | Up to 4,750
Model | vCPU | Memory (GiB) | Instance Storage (GB) | Network Bandwidth (Gbps) | EBS Bandwidth (Mbps)
-- | -- | -- | -- | -- | --
c5.large | 2 | 4 | EBS-Only | Up to 10 | Up to 4,750
c5.xlarge | 4 | 8 | EBS-Only | Up to 10 | Up to 4,750
c5.2xlarge | 8 | 16 | EBS-Only | Up to 10 | Up to 4,750
https://medium.com/pinterest-engineering/improving-distributed-caching-performance-and-efficiency-at-pinterest-92484b5fe39b
Most instances are r5 ones:
![image](https://github.com/voedger/kb/assets/11589750/fb0e47ef-4e9c-41cc-aab9-525e32e6037e)
Instance Size | vCPU | Memory (GiB) | Instance Storage (GB) | Networking Performance (Gbps) | EBS Bandwidth (Mbps) -- | -- | -- | -- | -- | -- r5.large | 2 | 16 | EBS-Only | Up to 10 | Up to 4,750 r5.xlarge | 4 | 32 | EBS-Only | Up to 10 | Up to 4,750 r5.2xlarge | 8 | 64 | EBS-Only | Up to 10 | Up to 4,750 Model | vCPU | Memory (GiB) | Instance Storage (GB) | Network Bandwidth (Gbps) | EBS Bandwidth (Mbps) -- | -- | -- | -- | -- | -- c5.large | 2 | 4 | EBS-Only | Up to 10 | Up to 4,750 c5.xlarge | 4 | 8 | EBS-Only | Up to 10 | Up to 4,750 c5.2xlarge | 8 | 16 | EBS-Only | Up to 10 | Up to 4,750180.000.000 / (6 cores * 5000) = 6000 rps per core