Bulkhead — walls between workloads
Anything workloads share, one of them can use up — so split servers, connection pools and threads into separate pools, and a flood or failure in one stays there. The flooded pool still fails, and the walls cost spare capacity.
12 minMediumStation: BulkheadMembers lesson
Membership isn’t open yet. The free lessons are open — 16 of them.
- Circuit breaker
- Flaky payments
- Bulkhead
- Load shedding
- Backpressure
What you'll be able to do
- 1Explain how separate pools of servers, connections or threads stop a failure or overload in one part from spreading to the rest.
- 2Predict which requests still succeed when one workload overloads a shared pool, and when it overloads only its own partition.
- 3Choose where to draw the partitions — by dependency, by workload priority or by tenant — and weigh the spare capacity each split costs.
Inside the lesson
- steps
- 10
- questions
- 7
- simulations
- 4
You guess first, then run the system and see if you were right.
Step 1 · The setup
Leg 1 was about failing well when a dependency breaks. This leg is about holding the line when one part of the system is flooded.
Checkout runs two kinds of work on one shop server: product search, and orders — where shoppers pay. On a normal afternoon that's 30 searches and 10 orders a second. The server is 45% busy, and an order takes about 5 ms.
Tomorrow a sale banner goes up. Searches will jump to 120 a second; orders stay at 10.
Steps 2–10 are for members
More on the Resilience Line
- Failure modes and availabilityOpen9 min
- Timeouts — capping the waitMembers9 min
- Retries — with backoff and jitterMembers12 min
- Circuit breakerMembers10 min
- Load shedding — turning the excess awayMembers12 min
- Backpressure — a queue that says noMembers11 min
- Rate limiting — a budget for each callerMembers12 min