Capacity Management – Where is my Crystal Ball when I need it?
This is a blog about capacity management (the reference to “crystal ball” to guide our future decisions), covering challenges, lessons learned and interesting topics. This time we will use the COVID-19 pandemic to illustrate the challenges of capacity management for IT in the enterprise.
I am not aware that any of our major customers had built contingencies into their 2020 IT budgets for the effects of a major pandemic (like COVID-19). While sometime in January, news of a novel coronavirus that was spreading in a province of China was beginning to emerge, its threat to the global economy was unimaginable and the impact on IT operations was unforeseen. But as the virus spread through Asia and then jumped to Europe and the Americas, many governments declared the State of Emergency, proceeded with strict lockdown measures that confined most of the country’s population and forced companies to rapidly improvise to continue to deliver access and services to customers and employees.
I still recall the first time in mid-March when I saw the following infographic in one of the daily televised briefings as a way to explain why hospitals were rapidly reaching saturation and why it was so crucial to take the drastic measures that were introduced. The spokesman for the Department of Health explained that if the rate of contagion could be substantially slowed down, the capacity in the healthcare system could adapt to absorb the increase in demand of new patients (and of course, continue to handle the rest of non-coronavirus emergencies).
Capacity management is the process we use to ensure that the business can cope in the most cost-effective manner with the demands placed on its resources (in the case of the hospitals, the resources can be the ICU beds, health care workers, respirators and other medical supplies) and its primary mechanism is to find a balance between supply and demand. In most cases, you will leave some extra capacity or headroom to handle small deviations in the expected demand (e.g., have spare capacity in the ICU beds to handle unforeseen emergencies). But the sudden and drastic changes in the demand curve that have been seen because of COVID-19 and the strict actions taken, have caused substantial capacity problems at many companies. Schools and universities transitioned to online education; the networks of the telecommunication companies were stressed by the increase in broadband internet access; the closure of retail stores and most bank branches drove a shift to online transactions; government sites were swamped with unemployment and economic relief claims; many companies realized they had insufficient VPN capacity and licenses to handle their home-based workers. These unplanned changes in the capacity demanded have forced companies to react quickly to increase the capacity available.
The relationship between supply and demand for scarce resources
Governments’ measures have had to simultaneously focus on reducing the demand while increasing the supply to cope with the exponential progression of the pandemic. In the case of hospitals, the rate of demand for admissions is directly related to the rate of contagion, which can be reduced by keeping as many people at home for as long as possible, restricting free movement, establishing “social distancing” and enforcing the mandatory use of face masks when in public. On the supply side, governments and health officials are rapidly increasing the capacity of rooms reserved for COVID-19 patients by delaying/reducing other surgical procedures and increasing bedding, medical supplies, doctors, etc. Countries were building new hospitals in record time and turning hotels and sports arenas into overflow medical facilities became impacting examples of the urgency required.
To know how much capacity is required we turn to Little’s Law (N = λR), which states that the average number of customers in a system is equal to their average arrival rate multiplied by the average amount of time they spend in the system. A COVID-19 patient with severe symptoms stays hospitalized for about 2-4 weeks (the “R” in the formula) before they have sufficiently recovered and released. We can see that with as little as 10 new daily hospital admissions (the “λ” in the formula) with an average stay of 15 days, we would have 150 COVID-19 patients actively treated. The number of required beds (plus the additional medical supplies, doctors, etc) would grow to 750 if the number of severely ill patients requiring hospital treatment reaches 50 each day. This illustration highlights the importance to reduce new transmissions to mitigate the risk of collapsing the healthcare system’s capacity to cope with the pandemic.
The essence of the IT capacity optimization process relies on obtaining reliable information about the expected workload change (e.g., the growth in webstore traffic resulting from a promotional campaign, the decrease in the usage of certain technology stacks as applications migrate to public clouds). The planner can then focus on key questions such as the following to arrive at a plan that can provide the required capacity when necessary and at the right cost:
- How much headroom do I need?
- What is the cost of not having enough?
- How much risk can I tolerate?
- What challenges or losses are introduced if we do not keep up with demand?
- How much capacity will I need and when?
- How fast can I add capacity?
- What are my options for extra capacity?
- What are the costs of those options?
As we gradually approach the “new norm” with countries beginning to lift restrictions, companies must develop strategies and procedures with the same rigor to optimize IT costs and mitigate business risk just as the healthcare professionals are doing to contain, slow the spread and soon, eradicate this global virus. Let us hope that our “crystal ball” can provide us advance notice of what is coming to be better prepared next time.
Want to learn more about the principles of capacity optimization?
Watch the on demand webinar Controlling Costs and Mitigating Risks in Uncertain Times hier
Erfahren Sie mehr:
More Capacity News: 2 Week’s free trial of Fusion’s Capacity Optimization Service