Scam or Savings? A Cloud vs. On-Prem Economic Slapfight
summary
Argues that cloud is usually chosen for speed and features, not lower cost, and that moving back on‑prem is rare except for small proofs of concept or companies that already run DCs; discounts kick in around large spend, break-even month four on a 36‑month Mac instance is real, and data egress kills you for heavy internet traffic; migrations incur huge people costs so people cost more than infra; conclusion: run the numbers for your exact workloads, not slogans.
AI-written summary · about
chapters (21)
- 0:00Introduction
- 1:49Not about spreadsheets
- 3:13Cloud discounts at scale
- 4:37Myth of cloud repatriation
- 7:15Data transfer costs
- 8:16People costs matter
- 10:03Repatriation index data
- 11:05Cloud default for workloads
- 12:49Cost not primary driver
- 14:14Long-term economic math
- 15:19Mac Mini case study
- 17:27On-prem for experiments
- 18:28Judging without context
- 19:54Hidden incentives
- 20:56Migration myths
- 22:40Choosing today
- 23:44When cloud makes sense
- 24:48Workloads fit on-prem
- 25:52VMware price hikes
- 26:13Key takeaways
- 27:16Gen AI disrupts economics
related talks
31mOn-Prem is the new Black
45mCloud Computing: Cloud in a cost-effective way
41mA cloud cost savings spectacular
26mWhen the Cloud s Reign is Over
20mAre the days of on-premise servers over?
33mThe elephant in the cloud: Strategies of cloud cost optimisation
37mHybrid OnPrem + Cloud Architectures
49mWhat to look out for when migrating your data to the cloud; is it REALLY the right choice?