It's A Trap!
summary
Argues that private clouds mostly underdeliver: IDC shows 70% of infrastructure spend stays on‑premise, DORA finds high performers are 24x more likely to use cloud capabilities, and Gartner reports 95% of data‑center attendees see private cloud problems. Explains failures: OpenStack effort, kubernetes’ lack of true multi‑tenancy, sunk‑cost decisions, and slower feature velocity than AWS (thousands of product launches), concluding managed public services usually win unless legal or specialized constraints force on‑premise.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 1:55Sam Newman
- 3:34IDC Data
- 4:52Public Cloud Spending
- 6:16Stalling Effect
- 7:29Kubernetes
- 8:13Devils Report
- 11:03OnDemand SelfService
- 14:19Broad Network Access
- 17:50Kubernetes Not Multitenancy
- 21:22Private Clouds Fail
- 26:17Public Cloud Features
- 28:43Private Cloud Features
- 32:59We Already Have Computers
- 35:26Reasons To Stick Out
- 45:47The sunk cost fallacy
- 47:57The evolution of private cloud
- 50:50Why use public cloud
- 52:12Hyperscalpers
- 56:03Questions
related talks
33mWhat the Cloud?
20mAre the days of on-premise servers over?
31mOn-Prem is the new Black
31mStay private or go public?
30mOn-premise data centers do not need to be legacy
50mA Cloudy Future or Can We See Trends?
38mIt's a Trap! How Abstractions Have Failed Us.
48mCloud Advisory: Cloudy, with a chance of people