Deployment Models at the Edge

Overview

The report is a 12-page briefing by T. Rahkonen, published by Uptime Institute Intelligence in October 2023 as UII Briefing Report 115. Its subject is deployment models for edge facilities and the IT they host. The report sits behind a registration form or an Uptime Intelligence subscription, so only facts from the public landing page and press release are used here.

The most visible finding on the landing page concerns workload character: 60% of workloads deployed at edge facilities are hybrid applications that rely on a central location. In many cases the edge therefore works not as a standalone mini data centre but as a layer that complements central infrastructure. That context shapes the resiliency and connectivity the edge actually needs. Small edge data centres are in the range of tens to hundreds of kW, and demand has fallen below initial expectations. According to the key points, high site-level resiliency expectations hold back smaller far-edge data centres because of opex and capex. At the same time, most workloads tolerate the added latency of metropolitan and regional facilities.

A regulatory dimension is included too: performance thresholds in regulation may cause some older edge sites to be scaled down or closed. Since the full report content was not seen, no comment can be made on its method or detailed figures. For designers its relevance lies in setting scale and resiliency expectations realistically for edge projects, planning remote monitoring from the start, and recognising that regulatory thresholds may put older sites at risk.

Key points

1. Published as UII Briefing Report 115 in October 2023; 12 pages.
2. 60% of workloads deployed at edge facilities are hybrid applications relying on a central location.
3. Small edge data centres are tens to hundreds of kW.
4. Demand for edge data centres is below initial expectations.
5. High site-level resiliency expectations hold back smaller far-edge sites because of opex and capex.
6. Most workloads tolerate the added latency of metropolitan and regional facilities.
7. Regulatory performance thresholds may cause some older edge sites to be scaled down or closed.

Using it in design

• Size edge templates in the tens-to-hundreds of kW range.
• Plan remote monitoring as a default; do not assume on-site staff at small sites.
• Factor the edge site's dependence on a central location into connectivity and redundancy decisions.
• Do not assume high site-level resiliency at every edge node; assess the workload's latency tolerance first.

Who should read it

Architects planning edge infrastructure, facility designers and managers scoping edge investments.

Publication details

Authors

T. Rahkonen

Publisher

Uptime Institute Intelligence

Year

2023

Document number

UII Briefing Report 115

Document language

English

Source type

Industry body

Access

Citation only

Licence

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

Citation