capacity planning
Estimating how much computing capacity a system will need and when, so it is there before the load arrives.
In more detail
Before the cloud this was a purchase decision made months ahead. In the cloud it becomes an ongoing job: choosing instance sizes, scaling bounds and quotas, and checking the limits of dependencies such as databases. Virtual machines leave more of it to you than managed services or functions do.
Example
copy_h = DATA_TB * 8000 / LINK_GBPS / 3600 # 2 TB over 1 Gbps is about 4.4 hoursRelated terms
- compute capacity — How much work a system can handle at once, measured in things such as servers, processors, memory or requests per second.
- Auto Scaling group — An EC2 feature that keeps a desired number of identical instances running, replacing failed ones and adding or removing them as load changes.
- service quota — A limit a cloud provider sets on how much of a resource one account may use, such as how many processors or networks per region.
- RTO — Recovery time objective: how long the business can accept a service being down before it is restored.
- recovery site — An alternative location where an organisation can run its critical systems and carry on working if its main site becomes unusable.
- bandwidth — How much data a network connection can carry per second, and by extension the amount of data a request or response uses up.
- BIA — A study of which business processes matter most and how the harm from an outage grows over time, used to set recovery targets.
- per node capacity — The throughput one server or node sustains while staying inside its latency target, found by a load test.
- headroom — Capacity kept free above the expected peak, so the system survives deployments, spikes and failures.
- failure domain — A group of components that can fail together because they share something, such as a rack, power supply or zone.
- utilization — The fraction of time a resource is busy.
Taught in
- Capital to Operating Expenditure — Cloud Infrastructure & Hyperscaler Architecture
- The Shared Responsibility Model — Cloud Infrastructure & Hyperscaler Architecture
- Comparing Compute Options — Cloud Infrastructure & Hyperscaler Architecture
- Cost, Budgets, and Tagging — Cloud Infrastructure & Hyperscaler Architecture
- Business Impact Analysis, RTO & RPO Modeling — Cybersecurity Fundamentals, GRC & Cryptography
- Disaster Recovery Sites: Hot, Warm & Cold — Cybersecurity Fundamentals, GRC & Cryptography
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture