RED for the service, USE for the disk

Watch 02

Choose

Photo pages read 64 KB documents at 8,800 reads/s. Users see p99 5.7 s and 10 percent errors. API CPU: 33 percent. Document store: CPU 15 percent, disk operations 18 percent busy, disk throughput 113 percent of its 500 MB/s limit. Pick the fix. Budget: 70 USD/mo.

Goals

Not tested yet

99 of 100 requests finish faster than this.The share of requests that fail or get no answer.What the servers and services cost each month.
Choose a design
Add 2 API servers
Double the document store CPU to 4 vCPU
Move the documents to a 1,000 MB/s disk volume