The Art of Capacity Planning: Scaling Web Resources in the Cloud

Copyright
Table of Contents
Preface
	Why We Wrote and Revised This Book
	Focus and Topics
	Audience for This Book
	Organization of the Material
	Conventions Used in This Book
	O’Reilly Safari
	Using Code Examples
	We’d Like to Hear from You
	Acknowledgments
Chapter 1. Goals, Issues, and Processes in Capacity Planning
	Background
	Preliminaries
	Quick and Dirty Math
	Predicting When Systems Will Fail
	Make System Stats Tell Stories
	Buying Stuff
	Performance and Capacity: Two Different Animals
	The Effects of Social Websites and Open APIs
	Readings
		Critical Section
	Resources
Chapter 2. Setting Goals for Capacity
	Different Kinds of Requirements and Measurements
		External Service Monitoring
		SLAs
		Business Capacity Requirements
		User Expectations
	Architecture Decisions
		Providing Measurement Points
		Resource Ceilings
		Hardware Decisions (Vertical, Horizontal, and Diagonal Scaling)
		Disaster Recovery
	Readings
	Resources
Chapter 3. Measurement: Units of Capacity
	Capacity Tracking Tools
		Fundamentals and Elements of Metric Collection Systems
		Round-Robin Database and RRDTool
		Ganglia
		Simple Network Management Protocol
		Treating Logs as Past Metrics
		Monitoring as a Tool for Urgent Problem Identification
		Network Measurement and Planning
		Load Balancing
	Applications of Monitoring
		Application-Level Measurement
		Storage Capacity
		Database Capacity
		Caching Systems
		Establishing Caching System Ceilings
		Special Use and Multiple Use Servers
	API Usage and Its Effect on Capacity
	Examples and Reality
	Summary
	Readings
		Performance
		Network
		Load Balancer
		Storage
		Database and Caching
	Resources
Chapter 4. Predicting Trends
	Riding the Waves
		Trends, Curves, and Time
		Tying Application Level Metrics to System Statistics: Database Example
		Forecasting Peak-Driven Resource Usage: Web Server Example
		Caveats Concerning Small Datasets
		Automating the Forecasting
		Safety Factors
	Procurement
		Procurement Time: The Killer Metric
		Just-in-Time Inventory
	The Effects of Increasing Capacity
	Long-Term Trends
		Traffic Pattern Changes
		Application Usage Changes and Product Planning
	Iteration and Calibration
		Best Guesses
		Diagonal Scaling Opportunities
	Summary
	Readings
		Buy or Lease
		Time–Series Forecasting
		Curve Fitting
		Measurement
	Resources
Chapter 5. Deployment
	Automated Deployment Philosophies
		Goal 1: Minimize Time to Provision New Capacity
		Goal 2: All Changes Happen in One Place
		Goal 3: Never Log in to an Individual Server (for Management)
		Goal 4: Have New Servers Start Working Automatically
		Goal 5: Maintain Consistency for Easier Troubleshooting
	Automated Installation Tools
		Preparing the OS Image
		The Installation Process
	Automated Configuration
		Defining Roles and Services
		An Example: Splitting Off Static Web Content
		User Management and Access Control
		Ad Hockery
		Example 2: Multiple Datacenters
	Summary
	Readings
	Resources
Chapter 6. Autoscaling
	The Challenge
		Autoscaling on Amazon EC2
		Design Guidelines
		Scalability Analysis
		Properties
		Autoscaling by Fixed Amount
		Scaling by Percentage
		Startup Time Aware Scaling
		Potpourri
		Advanced Approaches
	Summary
	Readings
	Resources
Appendix A. Virtualization
	Overview
		Looking Back and Moving forward
Appendix B. Dealing with Instantaneous Growth
	Mitigating Failure
		Graceful Degradation and Disabling Heavy Features
		Baked Static Pages and Beyond
		Cache but Serve Stale
	Handling Outages
Appendix C. Capacity Tools
	Monitoring
		Metric Collection and Event Notification Systems
		Ad Hoc Measurement and Graphing Tools
	Deployment Tools
		Automated OS Installation
		Configuration Management
		Cluster Management/Container Orchestration
		Inventory Management
		Trend Analysis and Curve Fitting
		Books on Queuing Theory and the Mathematics of Capacity Planning
Index