Efficiency as reliability practice: a shorter feedback loop

I care about “Efficiency as reliability practice” for one practical reason: it either leaves clearer ownership next week, or it was theater.

Flaky CI, chatty retries, and oversized environments are leadership issues because they burn attention and energy.

I will stay on this thesis — no adjacent manifesto, no tour of every neighboring discipline.

On Efficiency as reliability practice, refuse waste that buys no safety — retry storms, flaky CI, always-on idle fleets — and assign an owner who can change sizing or schedules. Unowned green metrics are decoration.

Keep the unit of progress small enough to finish under a full calendar: one owner clarification, one verification signal, one reversible control. Grand programs without weekly evidence become status machines.

A scene from delivery

Flaky tests burn CI minutes nightly. Engineers ignore failures. Carbon and attention both leak. Quarantine with a deadline beats living with noise.

That scene is the thesis under calendar pressure. The response is not another status meeting — it is a control.

Mechanisms

Architecture forums that never change a sizing or schedule decision are performing green theater. Require a reversible experiment with an owner.

Put sustainability questions in architecture forums: what did we choose that forces waste forever? If the forum cannot change a decision, it is theater.

Design for fewer retries, not prettier dashboards. Retries multiply compute and hide root causes. Cap amplification and alert when it exceeds a threshold.

Ephemeralize non-prod by default. Always-on demo and staging fleets are often habit, not requirement.

The energy cost of flaky CI includes engineer attention. Fix flakes as reliability work with an SLO on quarantine time.

Measuring compute waste without an owner who can delete or resize is dashboards as decoration. Pair every metric with a decision rights note.

Flaky CI burns minutes and attention. Quarantine with deadlines; infinite noise trains people to ignore real failures.

Another texture

A team celebrates resilience because the client retries aggressively. Production is stable-ish and the bill is not. They built a retry storm and called it fortitude. Cap retries and fix the root timeout.

Staging runs at production scale “for realism” while load profiles show it is idle 20 hours a day. Right-sizing is leadership because nobody else can say no cleanly.

Tradeoffs

Deleting pipeline stages requires courage and good eval of what still protects users.

Carbon-aware batch shifts can delay results. Make the delay explicit and reversible.

Right-sizing can feel like risk. Mitigate with scale-up paths and load tests, not eternal overprovision.

Efficiency is reliability practice. If a carbon or cost signal never changes an architecture decision, the forum is performing virtue.

Resist the urge to expand scope into neighboring slogans. If a control does not make Efficiency as reliability practice more operable for the next person, leave it for another note. Dilution is how coherent essays become stitched scrapbooks.

Failure modes

Each of these is a missing control, not a personality problem:

• Greenwashing metrics that never change an architecture decision.
• Always-on environments for demos that could be ephemeral.
• Flaky CI accepted as weather instead of a defect.
• Retry storms celebrated as resilience while they burn budget and hide bugs.

What to do this month

• Track flaky test quarantine time as a reliability SLO.
• Report idle environment hours next to cloud spend in engineering reviews.
• Ephemeralize non-prod by default; justify always-on in writing.

Retry amplification should have a budget and an alert. Unbounded retries are a distributed denial of your own reliability.

Right-size infrastructure as leadership work: challenge defaults that force always-on waste for rarely-used paths. “Just in case” environments need a written justification.

Right-sizing after a traffic shape change is a leadership conversation because finance, platform, and product all feel the “what if” differently. Bring numbers.

Measure compute waste in product teams: idle envs, redundant builds, chatty meshes with no owner. Numbers create permission to delete.

CI minutes are both money and attention. Quarantine flakes with deadlines; infinite quarantine is how noise becomes culture.

Efficiency as reliability practice shows up when you delete a redundant pipeline stage and incident escape rate does not rise. That is the experiment worth running.

Sustainable pace and sustainable systems share a refusal: do not buy speed with hidden churn — flaky tests, retry storms, or always-on idle fleets.

Carbon-aware scheduling without slogans means shifting batch work when it is safe and measuring whether load actually moved. Unmeasured virtue is marketing.

Architecture review asks about carbon once, records a slogan, changes no sizing decision. Next quarter the same waste remains. Sustainability reviews need owners who can change schedules or footprint.

A team discovers staging mirrors production capacity while serving a handful of QA users. Right-sizing frees budget and reduces the blast radius of bad deploys.

Carbon-aware batch jobs shift to night without checking downstream SLA commitments. Green intent creates customer pain. Constraints must be negotiated, not silently applied.

One more practice worth installing: Budget retries: cap and alert when amplification exceeds a threshold.

The point of writing this down is not documentation theater. It is so the next person does not have to reconstruct intent from Slack.

Close

Refuse waste that buys no safety. That is both green and kind to the on-call.

If someone new cannot explain Efficiency as reliability practice from your artifacts in fifteen minutes, the practice is still private.