Major sports and non-sports events have massive global visibility. They involve millions of spectators and visitors in the host cities and virtually they may reach billions. The event phase is typically short but represents the culmination of years of planning.
Very few areas are not touched by software, networks, and computing. Making this engine run smoothly to ensure amazing experiences for athletes, teams, participants, guests, spectators, visitors, media and the global audience is the goal of an effective technology operation.
And while every event is unique (multi-sport, single sport, world expo, festival, etc), the scope of technology operations is the mostly the same, with differences of scale.
What makes major events unique?
How hard can it be? It’s basically just ITIL/DevOps like any other project, right? In principle, yes – but there are things that deserve some extra attention.
From zero to complex
When a major event organising committee is setup many years before the event, there are very few systems in place, or stakeholders and end users to worry about. That will ramp up to more than 100 different technology services supporting several thousand staff and contractors, multiple stakeholder groups and millions of people relying on those technology services. Wrong decisions, poor processes or untrained teams may have very little impact in the early days of the project, but can be disastrous during event time when the scale, complexity and criticality peaks.
While establishing a common SLA across all technology services is essential, it is a minimum baseline. The reputation of the host country is at stake as is the brand of the chosen technology partners. SLA service credits may help to focus attention on quality of service but the important criteria are less tangible. Were there any technology issues visible to the public? Have any technology issues prevented the successful operation of the event? How do the key stakeholders perceive the services provided by the Technology team?
A diverse team with a shared goal
The world’s largest events draw on a talented workforce of individuals and companies from the host country and from overseas. This involves different national cultures, company cultures and varying levels of experience. During event operations they must work as a single team using common processes, tools and means of communications. It’s easy to work in silos and blame the other guy. Getting a diverse set of technology partners, providers and team members to work as a single team with a single goal is the best way ensure a successful outcome.
Rapid team ramp up
Just as the services ramp up from zero, so does the technology operations team. But this is far more concentrated on the final few months leading up to the event. The last few months see hundreds of staff, contractors and volunteers join and take their operational role. They need to be identified, assigned, trained quickly and made effective in their operational role. Even those who have been working on the project for some time will transition to an operational role which may be very different to their project role.
The scope of technology operations
By operations, we refer to the time a technology service has gone into live use for end users. Once it reaches this production state both the end users and the service itself needs to be supported to ensure that it meets the expectations and service levels. This may be called service management, service delivery management, production operations or run services. For the purpose of this document we call this technology operations.
Note: It is common that in the early stages of the project, business owners don't have clear requirements. Then Agile delivery via sprints is the best way to progressively deliver functionality.
Key operating principles
Those who build it, support it
The team that develops the applications should support them during the event. The team that installs and configures the network should then monitor it, resolve issues and manage changes. It's not practical or effective to handover support to another support team who have limited knowledge of the project. As the event has a limited duration, better quality is achieved when the team that builds the services, supports them. This aligns with devops principles.
Support ownership via an accurate technology service catalogue
The technology service catalogue should provide clarity on all characteristics of a service, including who is supporting its different components. Quick access to the support team information for each service ensures that the right people are rapidly engaged in support issues.
Ownership AND teamwork
Ownership for the support of technology services and their components is essential, however it must be complemented by a one team approach to avoid creating silos of competence. That includes the organiser staff, technology partners, suppliers and other stakeholders.
Go operational as early as possible
Working as a single team with common processes, tools, organisation model and KPIs is easier when the team and number of operational services is small. It is much harder to transition a full project team to an operations team only a few months before the event.
Transparency / no blame
To resolve major incidents or address their root causes, it requires transparency and full information sharing. In many cases human error will be a cause or a contributing factor. By identifying all root causes, including human error, they can be addressed so they don't happen again. To encourage transparency a no blame culture is essential. People will make mistakes, but the team will be better off having learned from those mistakes.
Continually create and reuse knowledge to get better
Preparing a major event is a learning journey. The goal is for the services and the team to be 100% ready when it starts. Applying lessons learned throughout the project is an effective way to accelerate the team competence.
Technology operations framework
The following diagram outlines the core components of technology operations for a major event
This guide uses some common industry standards as a basis.
· ITIL - Information Technology Infrastructure Library
· SRE - Site Reliability Engineering developed by Google
· DevOps - a set of practices that combines software development and IT operations
Terminology is often event specific. For example, some events may use the term Technology Operations Centre while another may call it an IT Command Centre, but the functions are very similar. Terminology will also evolve over time. For this guide we have used either the most common term or the one we thought made most sense at the time.
Mistakes, changes and improvements
The aim of this guide is for it to be useful and provide knowledge that others can build upon. If you would like to improve this guide for others, please let us know.
The guide reflects the knowledge of the Alkira team across many years and types of events. Many others have contributed to the evolution of technology operations concepts. We hope that this body of knowledge can continue to expand, improve and be beneficial to all.