BDMAAS: SMART AND BUSINESS-DRIVEN AUTOSCALING FOR LARGE-SCALE CLOUD BASED IT S...
summary
Claims that BDMaaS (an open‑source JRuby-based engine) actually evaluates alternative VM placements and instance counts against real AWS pricing and SLO penalties, uses a simulator plus optimizers (genetic algorithms and particle swarm optimization), and finds a monetary “sweet spot” across scenarios (e.g. 5%/25% workload shifts) rather than relying on CPU thresholds. Implementation exists, is validated on distributed AWS setups, produces measurable cost vs. penalty tradeoffs, and is shared as runnable code.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:55Be openminded
- 3:02AWS
- 8:29Business Impact Analysis
- 10:22Scaling Too Much
- 12:08What is Scaling
- 13:34What is Analysis Simulation
- 13:55Business Tree Management
- 14:32Example
- 18:50Optimization
- 22:43Assumptions
- 25:07Service Provider
- 26:36Workflow
- 28:52Organization Solution
- 30:02Simulation
related talks
45mAutoscale — Pay When You Earn
45mAutonomous Multi-Cloud Serverless Deployment and Optimized Management
27mMastering AWS Cost Optimisation
31mAccelerate AI adoption: automated cloud efficiency
23mProactive and Polymorphic Adaptation of Multi-Cloud Deployments
40mHow an AI driven approach reduces cloud cost and makes your Kubernetes infrastructure autonomous
31mAutoscaling Apps without Supersizing Costs
28mScale Smarter Not Harder: How Extending Cluster Autoscaler Saves Millions