Tag: cloud

Posted on

Upcoming articles: September 2024

I am writing the following pieces, to appear in Computer Weekly in September.

Please note the earlier than usual deadlines, due to the Bank Holiday weekend.

How to succeed at cloud repatriation 

Deadline for input: 1700hrs, Friday 16th August

In this feature, we will look at how to make cloud repatriation work. This will include:

  • Which data are best suited to cloud repatriation?
  • Which workloads (and applications) benefit most from moving back on premises?
  • How should you prepare your private infrastructure tfircloud repatriation? What are the potential pitfalls? What could you overlook?
  • How do you make your data / infrastructure future-proofed if you repatriate from the cloud? How do you ensure you will remain cloud-native
  • How do you ensure you can reverse the decision and smoothly transition to the cloud, if the need arises, or use it for burst use cases?

I am looking for CIOs, analysts or consultants’ views for this piece.

What you need to know about Kubernetes DR 

Deadline for input: 1700hrs, Wednesday 21st August

With Kubernetes becoming more widely used in enterprises, the question of how to ensure it can survive an outage or systems failure is ever more critical. We’ll look at the points below; note this is different to K8 backup, which we have covered before.

  • Why do we need DR for Kubernetes?
  • What are the challenges to doing DR for Kubernetes clusters? 
  • What are the infrastructure requirements of DR for Kubernetes?
  • What are the risks to Kubernetes environments that need to be mitigated by DR; how do these differ from other environments)?
  • What key points would a DR plan for a Kubernetes environment contain?
  • What kind of products can help with Kubernetes DR? 

Again the preference is for analyst and consultant views, not vendors.

Data centre cooling

For this feature, we are looking into the state of the art for data centre cooling. What ticks the boxes for effectiveness, efficiency, cost and energy use? Which solutions are easiest to fit into current data centres, and which improvements bring the greatest benefits?

The rise of new workloads like AI means that datacentre operators need to consider how to get more power in, to support large GPU clusters, and the cooling requirements of these chips.

So how will operators tackle those new power and cooling requirements?

The focus will be on the technologies currently on the market, and which can be applied to existing data centres. We can touch on emerging technologies and those that work with new-build locations, but the focus of the feature is on practical options CIOs and data centre managers can buy now.

I am keen to hear from data centre owners/operators — including CIOs — industry consultants and analysts. Due to the nature of the feature, we are unlikely to quote equipment vendors directly, but happy to receive information on their solutions.

To contribute to any of these pieces, please contact me by email in the first instance.

Posted on

Upcoming articles: Computer Weekly, April 2024

I am working on two storage-related features for Computer Weekly, with April deadlines.

When is cloud data storage the answer?

Has there been a cooling off in enterprise enthusiasm for cloud storage?

In this piece we will look at some reasons firms should move their data storage to cloud infrastructure.

The piece will follow up on this from earlier in the year: Cloud is not always the answer: Five reasons why

We aim to cover

  • When does on site storage fall short and how cloud can be better?
  • Performance / availability / operations:
  • IT management:
  • Costs
  • Agility and responsiveness
  • Data protection and compliance

The deadline for leads is 1700, Thursday 28 March.

How does AI affect data storage

The second looks at the impact of AI workloads on storage. This includes how data are used in training models, how AI analytics are used, where data are sourced and where they are stored. The piece will also cover compliance, which is a growing concern for anyone using AI.

The piece will not be limited to LLMs or generative AI, but all the different forms of AI being used by enterprises.

On the technology side, we’ll cover IO, the types of storage being used, where the data processing bottlenecks sit, how to feed GPUs and where to store the outputs from AI systems. But that’s not the whole picture.

I’m looking for analysts and consultants with direct experience of AI projects.

Deadline for leads: Wednesday 10th April, 1700hrs.

To contribute to either piece, please contact me by email in the first instance.

Posted on

Computer Weekly features: January 2024 publication

I am writing two features for publication in January 2024. Note that due to early filing dates, the research needs to be completed by December 23rd, 2023.

Feature 1: Cloud to Cloud backup

This piece will look at the shortcomings of native cloud data protection, and their impact on both operations and compliance. It will look at cloud to cloud backup as an alternative, examining:

  • how it works
  • the workloads it is suited to (and any performance impacts)
  • who provides cloud to cloud backup services

Feature 2: Reasons not to use the cloud

Covering data storage, primarily, the piece will explore reasons why an organisation might choose not to use the cloud. This cloud include, but is not limited to:

  • operational needs and performance
  • IT management difficulties
  • costs
  • data protection and compliance
  • issues making cloud fit the business needs

For both pieces, happy to look at white papers and analyst research as well as consultants’ and CIOs’ views. As ever, it’s best to email pitches or input in the first instance.