Event Recap

RECAP | ROUNDTABLE | Data Center Infrastructure and Sustainability

Introduction: Uptime Institute lists Sustainability as one of the big trends and challenges for 2020. Data center operators have claimed advances based on small, incremental, and relatively inexpensive steps. However, the times of easy wins is ending. Regulators, watchdogs, and customers will increasingly expect digital infrastructure operators to provide hard and detailed evidence of carbon reductions, water and power savings – all while maintaining 100% uptime and resiliency. Terry Altom, Uptime Institute Senior Consultant, attended to contribute to the conversation.

What are the primary methods you are attacking to reduce your data center carbon footprint? Do you have a 5-year plan on how you intend to reduce water and power usage? What are the key performance indicators being utilized to track improvement? What tools are being used to actually measure and report improvement?

Roundtable discussion:

In this roundtable, members discussed what they are actually doing around sustainability. The attendees indicated they are still focusing on PUE (power usage effectiveness) from a data center perspective. In this regard, it was mentioned PUE is the perfect metric for data center infrastructure improvement, but not perfect for tracking and comparing data centers against each other. For the most part, operators have addressed the items referred to as "low hanging fruit” to improve PUE:
• Eliminating bypass airflow in the computer room
• Hot/cold aisle and containment in the computer room
• Economization with cooling systems
• Focus on UPS system losses and cooling optimization
• Computer room supply air temperature and set points
• Address zombie servers

All these low hanging fruit items should be considered best practices in new data center design. Next steps require investment. One attendee mentioned their bigger focus is now on equipment being life-cycle replaced. They are looking at newer more efficient technologies that utilize free cooling and economization in various climates, more efficient UPS systems and batteries, as an example. Terry Altom chimed in that the first cost should not be the driver when purchasing new equipment. The total return on investment (ROI), taking efficiency into account, needs to be considered. It was also mentioned how micro-grids is a technology being targeted in order to ultimately get to carbon neutral.

Beyond the data center infrastructure equipment, attendees talked about how IT hardware needs to be included in the efficiency discussion. IT hardware needs to have highly efficient power supplies, virtualization needs to be utilized to increase utilization, and server power management needs to be utilized to improve idle consumption. Also, better server rack management needs to be deployed to track and manage power and cooling consumption (ex., smart rack mounted power distribution, rack sensors for server intel cooling).

The conversation then shifted to water usage. A number of the attendees indicated they are tacking or starting an initiative to track WUE (water usage effectiveness). One attendee commented that “water is the new oil”, and because of this water usage is getting heavily scrutinized. Attendees indicated they are looking heavily at air cooled chillers, evaporative cooling, and DX systems as viable cost-effective alternatives to utilizing makeup water. An attendee also mentioned how they are checking into rainwater harvesting to save some money and increase redundancy for makeup water where applicable.

The attendees were then asked whether they have a plan to reduce carbon. Many of the attendees indicated their companies have announced they intend to become carbon neutral within a certain timeframe (ex., by 2030). They are finding this commitment to carbon neutrality is freeing up money for investment, and they are pulling together estimates and evaluations to demonstrate the cost and the carbon reduction return. This is also aiding in the money available for life-cycle replacements that focus on improving equipment efficiency. The comment was made that once management realized efficiency and reliability were not exclusive, there has been more focus toward spending money on improvements. Examples of other technology opportunities being tried and looked at are the use of fuel cells, liquid immersion cooling for servers, and running UPS systems in eco mode. It was mentioned ASHRAE 90.4 encourages designers and owners to incorporate waste heat recovery as an example. All these items are being considered because a sound economic business case can be made, they provide a positive impact toward carbon reduction, and there is no impact to reliability.

Lastly, the use of Cloud was mentioned. One of the rationales of public cloud is sustainability and could be considered a strategy for moving to carbon neutral. Companies essentially shift the responsibilities by moving IT out of their portfolio. However, it was pointed out that Cloud still should be tracked by the company because the reality is the company procuring Cloud services still needs to be accountable (i.e., account as Scope 3 and therefore still responsible).

Request an evaluation to view this report

Apply for a four-week evaluation of Uptime Intelligence; the leading source of research, insight and data-driven analysis focused on digital infrastructure.

Posting comments is not available for Network Guests