Skip to header Skip to main navigation Skip to main content Skip to footer
Computernik Networks
Remain at the cutting edge of technology

Main navigation

  • Home

Half of Enterprises Run Workloads in the Wrong Place, Akamai Study Finds

Server racks in a computer server room, representing the cloud data centers where enterprise workloads are placed.

Half of enterprise IT leaders are knowingly running workloads in the wrong place, according to a new study released on November 16 by Akamai Technologies. The report, titled Right Workload. Wrong Place, was based on a commissioned survey conducted by Forrester Consulting of 528 enterprise infrastructure decision-makers across North America, EMEA, and APAC at companies generating 250 million to over 1 billion dollars in annual revenue.

The central finding is not that companies chose poorly at the start, but that they feel trapped afterward. Fifty percent of organizations said they have deliberately left workloads in nonideal locations simply to avoid the cost and effort of moving them. Once a production workload lands in a cloud region, nearly eight in ten respondents said relocating it takes days or longer, 38 percent said it requires re-architecting, and 8 percent said they cannot move workloads at all once deployed.

That rigidity is colliding with a workload mix that has become far more interactive. Three-quarters of enterprises, 76 percent, said interactive or real-time workloads now account for more than 25 percent of their total software portfolio, and 40 percent said those applications represent more than half. Yet cloud placement has not kept pace. Fifty-one percent of the organizations surveyed run production compute in five or fewer regions, and 37 percent admitted their infrastructure is centralized because of provider constraints rather than a deliberate architectural choice.

The performance cost is measurable. Eighty-two percent of enterprises said they miss response-time targets for business-critical interactive workloads, with 44 percent failing on 5 percent or more of all requests. Respondents pointed first to data separation, with 41 percent citing retrieving data from a different location than where it is processed as the primary driver of response-time issues, followed closely by the physical distance between compute and end users at 38 percent.

Money reinforces the trap. Eight in ten respondents, 80 percent, reported unpredictable data movement costs. Sixty percent said egress and transfer costs dictate where they store data relative to compute, and 54 percent said those fees force compromises on caching and replication strategies. In other words, the pricing model that makes moving data expensive also makes keeping it in the wrong place rational, at least in the short term.

The business consequences over the past 12 months were concrete. Fifty-five percent of respondents suffered performance degradation during regional traffic spikes, 44 percent overprovisioned unused capacity, and 34 percent were forced to delay product or feature launches. Akamai Vice President of Cloud Product Marketing Ari Weil said enterprise infrastructure is reaching a turning point, arguing that running compute far from where data lives creates delays and inefficiencies that organizations can no longer sustain financially, operationally, or strategically.

Asked what they would do without provider or complexity constraints, 75 percent of IT leaders said they would deploy compute in more geographic locations, and 39 percent said they would at least triple their current physical footprint to achieve faster response times and enter new markets. For enterprise architects, the message is practical: placement decisions made for real-time APIs, AI inference, and interactive services deserve the same scrutiny as compute selection itself, because the study suggests the hardest part is not choosing a region on day one, but being able to change that choice later.

Photo: Server racks at the NOIRLab Headquarters Computer Server Room in Tucson, Arizona. Credit: NOIRLab/NSF/AURA/T. Slovinsky, Creative Commons Attribution 4.0 International License, via Wikimedia Commons.

Cloud Computing
Enterprise IT
Data Centers
Edge Computing
IT Infrastructure
Artificial Intelligence

Visit Red River New Mexico

Visit Our Vacation Website, Red River New Mexico

Copyright © 2026 Rocky Mountain Madman LLC - All rights reserved

Developed and Designed by Rocky Mountain Madman LLC