Uptime Institute 2019 Global Data Center Survey (9th annual) - press release

Overview

This document is the press release announcing the ninth annual results of the Uptime Institute Global Data Center Survey for 2019. The survey ran in March-April 2019 and drew on nearly 1,100 end users in more than 50 countries. The release gives a broad snapshot of the industry, concentrating on where corporate computing still runs, why outages happen and how hard it is for operators to find and keep skilled staff.

The first headline message is that enterprise-owned data centres continue to host more than half of all workloads. Forty percent of respondents said they use availability zones, meaning at least two active data centres replicating each other, to achieve resiliency. This points to a picture in which redundancy inside a single facility is no longer regarded as sufficient on its own, and where resilience across sites is becoming common practice.

The outage findings are the most instructive part for designers. Sixty percent of respondents said their most recent significant outage could have been prevented through better management, processes or configuration discipline. More than ten percent reported that their latest reportable outage cost over $1 million. Power loss accounted for one-third of reported outages, while aggregated IT-stack problems accounted for twice that share.

The survey also recorded a rise in average PUE to 1.67 and found that 61 percent of respondents face significant difficulty recruiting or retaining staff. The figures reflect self-reported user answers and are not a substitute for measured field data.

Key points

1. Survey conducted March-April 2019 with nearly 1,100 end users in more than 50 countries.
2. 40% use availability zones (at least two active, replicating data centres) for resiliency.
3. 60% said their most recent significant outage could have been prevented with better management, processes or configuration discipline.
4. More than 10% said their most recent reportable outage cost over $1 million.
5. Power loss caused one-third of reported outages; aggregated IT-stack issues caused twice that share.
6. 61% report significant difficulty recruiting or retaining staff.
7. The 2019 survey recorded average PUE rising to 1.67.

Using it in design

• In reliability analysis, weight on-site power as the single largest outage cause and assess redundancy (UPS, generators, distribution) accordingly.
• Alongside redundancy investment, list operating procedures, configuration discipline and change management as outage-mitigation measures in the design report.
• Use the availability-zone prevalence figure as context when weighing multi-site architectures against single-facility redundancy.

Who should read it

Data centre owners, operations managers and engineers designing for reliability who want a short industry-pulse reference.

Publication details

Authors

Uptime Institute

Publisher

Uptime Institute

Year

2019

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