The cloud has a people problem

The cloud has a people problem

In tech, we like to pretend that every problem can be solved by throwing hardware or software at it. We assume insights will magically appear from data if we apply the brute force of algorithms and servers. We believe applications can blithely fly from cloud to cloud through “single panes of glass” and portability layers. In sum, we forget that technology is really about people and what they do with that technology.

For those who have chosen to forget this truth, Gartner analyst Lydia Leong offers 1,172 words of reproof that the people problem applies to cloud. Anaconda’s survey of data science professionals rings the same warning bell for data science, with 90% of respondents acknowledging concerns over the potential impact of a talent shortage, and 64% expressing concern about their company’s ability to recruit and retain technical talent. Even in the land of abundant open source, we have a people problem, where concerns over sustainability should be focused on caring for individual project maintainers.

So, people matter. In the cloud world, recognizing this fact can help us be more pragmatic about how we approach cloud strategy and the people that implement it.

Multicloud reality

Much of the multicloud marketing in the past several years has had more to do with what vendors wanted to sell than what customers could realistically implement. We spent far too long talking about a single app working seamlessly across multiple clouds and not nearly long enough thinking through who was capable of building such a thing. Yes, specific software-as-a-service vendors can take care of all that underlying complexity (eg, compute in one cloud is very different from compute in another), ensuring your data layer, for example, is consistent across clouds. You can also shift to a microservices-based architecture, as Snap haswhich can make it feasible to pick and choose different clouds to host specific services.

But let’s not kid ourselves that this is simple. In fact, as a System Initiative CEO Adam Jacob points out, “Multicloud as a strategy for a single app appears to be dead/dying. Instead, it’s ‘choose the right cloud for the application’ based on what can be rearchitected.” This isn’t always true (as mentioned, SaaS providers can deliver multicloud for a single app by taking care of the data layer or other aspects of the application), but it’s much closer to reality than the old marketing of multicloud, wherein apps magically ran across multiple clouds. As Aiven’s Ian Massingham notes“Customers didn’t want to spread applications across multiple providers, but legacy vendors really, really wanted them to think that this was a good idea.”

In this rising awareness, Jacob goes on, “I don’t need portable plumbing. I need a better way to adopt and manage the rearchitecture” to take care of “specific components in unique software architectures.”

Copyright © 2022 IDG Communications, Inc.

Source link

Leave a Comment

Your email address will not be published. Required fields are marked *

%d bloggers like this: