project
AWS block storage; cited as the reason stopped instances remain a billing problem alongside retained Elastic IPs.
No evidence-backed relationships are recorded.
AWS's AgentCore Runtime Instances put three agents and 18 GB of models on one 24 GB A10G host with a shared, persistent disk. The design cuts cold starts and S3 handoffs on a single host; teams still budget GPU memory themselves and pay for the hours the host sits idle.
Reality
ClickHouse records a disk for every part, so a TTL rule can age old months onto S3 with the queries untouched. The lab behind the write-up substitutes MinIO for S3, so the latency half of the trade is still unmeasured.
Reality
AWS EBS volumes are zonal, so a StatefulSet pod can only run in the zone that holds its disk. An admission policy reads that zone and places eligible stateless pods beside it, for a claimed saving above 10%.
Reality
A dev.to writeup puts 30-50% savings in a cleanup phase most teams skip. That number is a share of wasted spend, not of the invoice, and the arithmetic matters before you promise finance anything.
Reality
A dev.to breakdown pairs each idle AWS resource type with a measurement window and a monthly list price. Eleven such resources add up to about $1,016 a month.
Reality