Operators face wider range of resiliency challenges: Uptime Institute Data Center Resiliency Survey 2023 (UII Data Report 99)

Overview

UII Data Report 99 is a six-page public summary of the resiliency survey that Uptime Institute Intelligence ran in January 2023. The survey was conducted online and reached more than 750 respondents. Because the report is short, its charts concentrate on a few core questions: who performs risk assessments, which areas they cover, what caused outages over the past three years and what is happening with insurance.

On risk assessment, 89 percent of respondents conduct their own assessments, and 89 percent of those repeat them at regular intervals. Coverage varies by area: electrical systems are assessed by 79 percent, mechanical systems by 72 percent, climate risk by 47 percent and sustainability by 46 percent. The gap between electrical and mechanical topics on one side and climate and sustainability on the other is notable.

The headline finding in the outage-cause section is that network and connectivity problems are the most common cause, at 31 percent. Power-related causes come second at 23 percent, and cooling is far behind at 2 percent. According to the report, network outages are usually less severe than power failures. For major network outages, the top cause is configuration or change management failure, at 45 percent (n=174).

The limits are clear: this is a short data report, so it offers no causal analysis or detailed discussion of method. Even so, it suggests that the traditional design focus on electrical and cooling redundancy should be weighed together with network and process risks, which account for a large share of outages. The results reflect the general experience of surveyed operators, not any single facility.

Key points

1. The survey was run online in January 2023 with more than 750 respondents.
2. 89 percent conduct their own risk assessments; of those, 89 percent do so at regular intervals.
3. Assessments cover electrical 79 percent, mechanical 72 percent, climate risk 47 percent and sustainability 46 percent.
4. Network and connectivity issues are the most common outage cause (31 percent), power-related is second (23 percent), cooling only 2 percent.
5. Network outages are generally less severe than power failures.
6. Configuration or change management failure is the top cause of major network outages (45 percent, n=174).

Using it in design

• Add climate and sustainability risk categories, which operators typically omit, to the tool's risk-assessment checklist.
• In resilience reviews, treat network and connectivity redundancy as its own heading, not only electrical and cooling redundancy.
• Ask for change and configuration management procedures as part of design handover documentation.

Who should read it

Engineers who run risk assessments and managers who decide on resilience investment.

Publication details

Authors

D. Donnellan; A. Lawrence

Publisher

Uptime Institute Intelligence

Year

2023

Document number

UII Data Report 99

Document language

English

Source type

Industry body

Access

Free to read

Licence

Copyright retained; citation only — the publisher’s terms apply to the original document

Citation