Why Operations Won’t Die: The Real Role of Ops in the Cloud Era

What is Operations

Operations is a function that emerged with the software industry and will only disappear when the software industry does; people may vanish, but functions persist.

From the whole software engineering perspective, we need people for requirement breakdown, development testing, and also deployment and maintenance, which gave rise to operations.

The software lifecycle resembles human life, but development only handles birth, not nurturing.

Who nurtures? Operations.

Developers often say, “It runs fine in my environment…”.

Your development environment runs alone, while production may have tens of thousands or even millions of concurrent users.

A colleague once asked whether architects should come from the development department. The answer: architects should actually come from operations.

Consider the three “highs” – high availability, high performance, high scalability – they all require a production‑centric perspective.

Operations Functions Beyond Deployment

In more than twenty years, operations has evolved into four recognized responsibilities: quality, efficiency, cost, and security (many companies’ security departments are overseen by the ops director).

Operations should thank cloud computing.

First, cloud computing reduces the chance of being scapegoated and automates resource delivery.

Second, cloud forces ops to improve and deliver higher‑value business work instead of remaining mere craftsmen.

Cloud computing is a new silo that intensifies quality and security challenges, making operations even more critical.

If business and system disaster recovery are not performed, incidents such as cloud‑provider outages or I/O hangs leave ops helpless.

Who coordinates? Only operations.

Efficiency is more of a business concern and is not tightly linked to cloud infrastructure.

Cost‑wise, cloud services are more expensive than amortized physical machines over three years; adding redundancy can double the expense.

One of ops’ values is saving money: Tencent’s SNG operations team saved 1 billion RMB per year through backend optimization.

But saving money is only a symptom; the core is ops’ deep mastery of technology and business, a genuine love for the company.

Means to Achieve Functions May Change

Ops has progressed vertically from manual (hand‑crafted) operations to automated and then intelligent operations, following natural development laws.

Horizontally, the industry moved from automation to DevOps, which also follows the same pattern.

Human society evolved from agriculture to workshop craftsmanship to industrial assembly lines.

The industrial era splits into mass‑production Ford lines and lean‑production Toyota lines.

The software industry was born after industrialization, but it is still in a painful workshop stage.

Programmers calling themselves “code farmers” is not just self‑deprecation.

The software industry must obey existing laws, learn from industrialization, and build pipeline‑style production.

Agile development, popular in software, originates from the lean thinking of the industrial era.

However, agile only lets development and testing collaborate to quickly produce packages (the “baby”), without solving deployment and maintenance (the “parenting”).

Thus DevOps emerged; it is essentially an extension of agile toward operations.

But DevOps will not kill operations.

DevOps frees ops from routine, low‑value deployment tasks; some companies even let developers handle releases.

Good thing for developers – they can jump into the pits they dug themselves.

DevOps does not eliminate ops because ops’ rich functions—quality, cost, efficiency, and security—remain essential.

The boundary between development and operations may blur, but it does not mean one replaces the other.

Not Everyone Can Be an Operator

Only those who are cautious yet bold, responsible, and humble can succeed in operations. When failures occur and leaders are watching, poor mental resilience leads to fear.

Can Cloud Outages Kill the Operations Industry?

This question is almost laughable.

Cloud outages actually highlight the importance of operations; if a company believes external resources can solve all problems, it ties its fate to someone else’s belt.

Harsh truth: if you don’t die, who dies?

Both failing and non‑failing clouds cannot kill the operations industry for the same reason.

We hope media use their influence responsibly and avoid creating panic; the statement itself contains logical errors.

Operations’ Spring Is Near

You feel cold only because it seems winter.

Traditional enterprises are still stuck in manual operations, fiddling with script repositories.

The internet industry may only account for about 5 % of national GDP.

In the internet “red sea,” ops may feel undervalued, but traditional enterprises offer a vast “blue sea” market awaiting exploration.

The Internet of Things will further increase demand for operations.

Imagine every streetlight in Beijing becoming IT‑enabled; the operational workload would be massive.

Operations professionals need to upgrade their knowledge: learn Python for DevOps, explore AI for intelligent ops, and deepen business understanding to become technical operators.

The internal career flow of ops staff has nothing to do with whether the industry disappears. Ops need a sense of crisis, not fear.

The sky is as big as the world of operations.

Below are highlights from the Efficient Operations community discussion:

Ops internal job flow has nothing to do with the industry’s disappearance.

New‑era operators have a new mission; feel free to share your thoughts.

Click to read the original article and see the ops camel’s show time.