Capacity & Failover: Servers for a Target Utilisation, N + 1, Load After a Failure and a Chart
This lesson is part of the course
In this lesson
You size Wrenfield's pool for its peak with a target utilisation, add one spare server, calculate the utilisation per hour with all servers up and after one failure, the headroom that is left and chart both curves.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Round Robin: Request Number to Server with MOD, Requests per Server and Uneven Speeds Locked
- Weighted Round Robin: Weights, Expected Shares, a Weighted Cycle and Its Error Locked
- Least Connections: Open Connections per Step, Chosen Server with MIN + MATCH, Finishing Requests Locked
- Hash Mod N: Given Hash Values, Server = MOD(hash, N), Keys per Server and Imbalance Locked
- The Cost of Rehashing: From N to N + 1 Servers, Keys That Move and Why It Is So High Locked
- Consistent Hashing: Servers and Keys on a Ring, Next Server Clockwise and Keys Moved When a Server Joins Locked
- Virtual Nodes: Several Positions per Server, Arcs of the Ring, Load Spread and How Many to Use Locked
- Capacity & Failover: Servers for a Target Utilisation, N + 1, Load After a Failure and a Chart Locked