RECAP | ROUNDTABLE | Climate Change – how are you preparing for extreme weather?

This Roundtable: Climate Change – how are you preparing for extreme weather? Looked at what advance considerations and planning should be put in place in face or increasing Extreme Weather events.

The Uptime Institute Consultant leading the discussion – Naveed Saeed, Vice President – Global Service Management – started the discussion by stating there are two ways to look at the topic – the facility itself and availability of essential services to the facility. Datacenters service many customers, and also rely on a number of external services to function.

So, to begin, Naveed recommended looking at how will the situation effect the supply chain? For example, if the facility is designed to have 12 hours of onsite fuel storage, consider increased the back-up supply and check with the supplier their own SLA’s in Extreme Weather conditions. If the ISP providing the Internet connection fails, the purpose of the facility is lost, so also consider checking their resiliency plans and incorporate multiple suppliers. Additional considerations would be for facilities using evaporative cooling, they will need to consider their water supply, onsite storage capacity and if this too could be extended?

ASHARE weather monitoring offices provides historic data points used as design data for extreme weather conditions, Climate Change has meant the surpassing of historical data points, and so Datacenter operators need to begin to look beyond them. The 1st step would be to look into the installed capacity of the existing footprint – chillers, cooling, DX Cooling or Cooling Towers. If the temperature goes 10-15% above the extreme design point for prolonged durations what will be actual capacity available to services critical functions. Extreme minimums also need to be considered; again, recent events have fallen outside of the ASHRAE data points. Extreme minimum can impact the equipment capacity to start to operate such as chillers, generators, pumps etc.

There should be a plan in place for Extreme Weather for all facilities, and it needs to be updated periodically based on updates risk register. A review cycle is needed to re-evaluate the risks, and Leadership and Management need to be involved to assign responsibilities. Part of the plan needs to be an understanding of the vendors proactive plan as well. On this point Naveed shared that he has recently been part of a conference organized by Coalition for Disaster Resilient Infrastructure (CDRI) in India where many in C-Level positions were keen to understand the risks they should be understanding with their Colocation providers.

A member based in the UK relayed an experience recently of Extreme Weather, in 2019 the temperatures varied between -12 to -39 degrees Celsius. This impacted the chiller systems especially, initially the facility team used emergency operating procedures, but needs long-term planning is in place to look at the design and operation of the Datacenters. Extreme low temperatures also effect backup time of UPS Batteries. When considering the equipment for low temps, running the generators more frequently, is also now a plan.

Naveed expanded on this to the Diesel on-site, is it the right grade? Is there diesel circulation system in place to avoid Gelling? Are the block heaters in place for the engine lubricants, especially for Extreme Cold Temperatures?

Another consideration from a member, based on experience, was to make sure that operations are planned for different facility load levels. In their experience a plan was in place for High Loads, but when the Extreme Weather hit the facility load levels were low and therefore the circumstances and outcomes were unexpected.

Bringing in experience from another industry, a member asked if Trace Heating could be implemented, even if it effected PUE? Naveed commented that availability of facility should always come before PUE.

Turning the Staffing Considerations, which have also featured prominently during the recent Pandemic, there were many experiences to be discussed. Measures currently in place include on-site sleeping arrangements and a call-around system. Concerns from staff regarding these plans varied from grievances from Unions to those that found it an adventure! It is also worth considering that staff might be unexpected trapped both on and off site. Those on-site will need food and water supplied, and for both sets of employees it is important that their own safety always comes before the availability of the facility.

So, in the face could the use of automation such as advanced BMS systems compensate for any lack of staffing or make remote operations viable? General opinion was that at the moment the high cost and security implications outweigh any benefits.
Final considerations from those on the call included:

• When moving to Renewable Energy Sources, their own reliability in face of Extreme Weather needs to also be considered and planned for
• A datacenter in a flood plain has ensured that they are responsible for supply of power to the Pumping Station nearby, therefore supply power from facility UPS system and placing the resiliency planning in their own hands

Wrapping up – the session looked to the future of what is possible – could helipads or drone technology be used to fly in supplies, or even staff during disaster situation?