On the last Wednesday Break celebrated at CAPSiDE, Enric Segarra highlighted, along other things, the importance of a company’s Sales team being capable of sensing or detecting paradigm shifts within the industry, and the company to support and bet on them for leading the sector in the mid or long-term. In the technology industry for example, the shift from the iron paradigm (physical servers stacked in data centres) to the cloud paradigm (virtual machines generated on-demand) is already being consolidated. Likewise, the operations team must detect the paradigm changes in their field and adopt them if they fit into the company and the team’s strategy and philosophy.

On Managing Your Self, Jagdish Parikh shows how things are changing in all aspects and levels: while strategies were planned before, now innovation is appreciated; old structures were hierarchical, while now it has an interconnected network shape; before, value was found in “being equal but better” than your competitors, now, differentiating factors are sought; the focus of attention is changing from the institution to the individual; and leadership no longer seeks to be dogmatic, but inspiring.

For the Operations teams, changes occur in 2 aspects:

Staying behind in any of both can have serious repercussions for the future of the company. For this reason, it has been spoken extensively about agile methodologies that have been gradually replacing the old methods for many years now. The changes agile methodologies imply range from CTO’s team management to the operators’ daily tasks.

In systems administration in particular, the norm until recently was the old-school model consisting of sysadmins working on separate projects individually, with task allocation by subjects (he’s an expert in databases, she’s an expert in web servers, the other one right there is an expert in computer clusters, etc.). This implies various problems, starting off with a very low bus factor, even up to type 1. This would mean that, if one of the team members left, the project would continue with serious problems or significant delays, with undesirable consequences for the company.

Agile methodologies push the team to communicate, to work together, to get everyone involved in all tasks, with a joint vision.

This doesn’t mean that all members of a team do the same job, but that they are aware of what their teammates do, in order to detect problems and solve them quickly. Everyone moves in the same direction. Agile teams are flexible, they adapt to unexpected events and have a high bus factor. Some of the most famous ones are KanbanScrum or Extreme Programming.

Although these techniques are often referred to software development teams, they are also perfectly applicable to system administrators. A joint meeting between the Barcelona sudoers group and the Comunidad Agile-Barcelona last March revealed that, after comparing experiences, the problems that affect ones and others aren’t as different as both usually think. It was also agreed that, even though it may look evident that new methodologies can help a lot to individual productivity and stress reduction, the resistance coming from some managers or group members made them hard to implement. However, it was concluded that it’s worth trying and insisting, because once you break this initial resistance, the advantages are immediately apparent.

Scrum’s stand-up meetings, the limitation of tasks that can be accepted with KanBan, or Extreme Programming’s Pair Programming are just some concrete techniques that can be applied. However, each team can adapt these techniques their own way and according to their normal functioning. If the team members are comfortable and the methodology application isn’t forced, the change is easier and everyone benefits from it.

In teams dedicated to technology, keeping up to date with changes in the industry is crucial to offer the best services in an agile and quick way, hence not tying the customer to a soon-to-be outdated solution and late responses to new needs. Having a flexible sysadmins team, willing to listen and to learn, capable of answering emergencies without causing damages to other clients and without experiencing difficulties in case a member is away, for the success of the IT department in any company.

At CAPSiDE we have been using agile methodologies in our “operations” team for a long time, and we try to improve and adapt them to the real needs of our engineers and customers. Thanks to them, we have noticed a significant efficiency and learning improvement within the team, which quickly results in a better and greater value for our customers.

TAGS: Bus factor, devops, Kanban, Methodology, Pair Programming, Scrum, sysadmin, sysops

speech-bubble-13-icon Created with Sketch.

Leave a Reply

Your email address will not be published. Required fields are marked *