Upcoming commission: Computer Weekly: Is there still life in spinning disks?

Conventional magnetic hard drives are being edged out by solid state storage across the enterprise. But there are still use cases where the spinning disk is best.

This article will look at the benefits of NVMe versus that of spinning disks, but also evaluate both technologies on cost, capacity and performance. It will then list the key applications and use cases where SAS or SATA drives are still the best options.

For this feature I am looking for analyst or consultants’ input, though case studies or white papers from vendors are welcome for background. Pitches by email by close of play, Tuesday 20 October – thank you.

Upcoming article: Unified File and Object Storage

For Computer Weekly, I’m looking at the emerging idea of unified file and object storage.

The piece will explain what this is, and more importantly why vendors are proposing it, and ask whether businesses should adopt it.

Are there use cases that could benefit unified file and object, and is this likely to become a significant segment of the market?

The deadline for suggested spokespeople is 1700 BST, Monday 5th October. I will then reply if further material is needed or to set up an interview. For this article, we are able to speak to vendors along with analysts and consultants/systems integrators.

As ever, contact me by email.

Upcoming article: why use tape storage today?

For Computer Weekly I am writing a feature on tape.

As a data storage medium, tape has been on the verge of obsolescence for decades. But the format endures. Why are IT and data managers continuing to choose tape?

The piece will look at:

  • The limitations and benefits of tape in today’s data centric environments
  • New and emerging tape formats and technology enhancements, such as software defined tape
  • How tape works with other storage media, including the cloud
  • Key use cases for the various tape technologies currently on the market.

The deadline to suggest interviewees or to share research is Wednesday, 5th August at 1700 BST. Initial submissions by email please.

Upcoming feature: Storage for AI, ML and analytics

For Computer Weekly my next feature will look at the specific demands placed on storage architecture by artificial intelligence, machine learning, and analytics.

The piece will ask:

What different approaches to providing storage are there for these technologies? 

What limits, performance considerations and bottlenecks exist with the different approaches?

What ways of providing storage for analytics are we likely to see in future?

The article will cover both on-premises and cloud-based storage, where relevant. I’m keen to include some real-world use cases if possible.

I am open to comment from industry professionals, consultants, analysts and CIOs working with AI. ML and analytics.

Deadline for leads: 1700hrs BST, Tuesday 23 June. As ever, please email in the first instance.

Upcoming articles: Compliance and Object Storage

I am writing two articles for Computer Weekly’s storage section, one on storage and data compliance for the enterprise, and the other on the growing field of high-performance object storage.

Data compliance

This piece will look at the top 5 UK compliance concerns in 2020.

What are the five key laws/regulations that must be adhered to by UK organisations in 2020, including both current and upcoming legislation. For each we will look at the implications of the law/reg for storage, backup, and archiving.

This could, for example, include legal search and e-discovery, or the Right to be Forgotten under GDPR.

We will also look at how the cloud fits in.

High performance object storage

Object storage has been known as a good way of storing lots of unstructured data, but with less emphasis on performance.

But AI and analytics workloads are prompting storage architects to look at performance too. The feature will cover:

  • Where object storage is heading in performance terms and what’s driving it.
  • Which performance metrics matter
  • How have object storage vendors improved performance?
  • Who are the key object storage vendors that are tackling the challenge of better performance and what do they offer?

The deadline for leads for both articles is Friday 20th March, please contact me by email if you can help.
.

Upcoming commission: computational storage

For Computer Weekly, I am investigating this emerging technology.

The piece will cover:

  • What is computational storage?
  • What is the architecture/key features of computational storage?
  • What use cases is it aimed at?
  • What are the pros and cons for those that might deploy it?
  • Who are the vendors and what products do they have?

I am looking for expert views – ideally independent — and real-world use cases or case studies. Deadline for leads: Wednesday 4th March, 1700 GMT. Drop me an emailif you can help.

Upcoming article: Flash storage tiers

For Computer Weekly, I am writing a short piece looking at the various tiers of solid-state storage.

What is on the market, how does each technology’s performance differ, and what are their applications? Why do businesses use tiered storage, and where does each solid state technology fit into tiering?

I will cover these storage types:

  • Optane/3D Xpoint/Z-NAND
  • NVMe
  • TLC Flash
  • QLC Flash

If you have market research or other information to share on these, the deadline to contact me — as ever by email — is 1700hrs, Friday 14th February.

Upcoming article: Who needs object storage?

This article, for Computer Weekly, will be an explainer on on-premises object storage. It will set out the key differences between object and block and file, and their pros and cons.

The article will ask:

  • What is object storage?
  • What are its key use cases?
  • What workloads need object storage?

The second half of the article will be a product section giving a vendor-by-vendor run-down on whether they provide hardware or software products, their architecture, scaling, speeds and feeds, data protection methods and other notable features.

There will also be a box out on cloud-based object storage.

Please submit background information such as white papers and case studies, product information, and suggested interviewees/experts, by 0900hrs Friday 19th July by by email in the first instance.

Please do not submit pre-written commentary or quotes.

Upcoming commission: unstructured data compliance

For Computer Weekly, I’m looking at the compliance issues around gathering, storing and processing unstructured data.

This article will examine the likely compliance risks in unstructured data, and suggest potential solutions. It will ask:

  • What is unstructured data? How does it compare to structured and semi-structured data types?
  • Why is compliance an issue at all?
  • Why is achieving compliance of unstructured data potentially problematic?
  • What are the key steps to achieving unstructured data compliance?

As businesses gather ever greater volumes of unstructured data, and develop new ways to process and analyse the information, compliance becomes increasingly important. This is especially the case when organisations start to combine data sets, and use advanced analytics to search for insights in the information. Does the original consent to hold and process the data carry over to this type of application? And what happens when unstructured data is mixed with other data sets?

For the piece I am keen to have comments from data scientists, compliance experts, academics, lawyers and end user IT organisations. As the deadline is quite short, please send pitches, initial comments and leads to me by 1200 London time, June 13th by email please.

Upcoming commission: storage for machine learning

Data storage is an often-overlooked part of machine learning and other AI deployments.

This article will appear in Computer Weekly. It will cover:

  • Definitions of machine learning/deep learning
  • Its storage requirements including
    • Sizing, capacity, performance (to match compute)
    • Scale
    • Media (SSD vs HDD, hybrids of the two)
    • Parallelism
    • Throughput vs IOPS
    • Locations – including use of the cloud

For this article we are open to comments from vendors, as well as analysts, consultants and other experts. Examples of ML use cases and how systems were designed to run it are most welcome.

Initial pitches and leads by Wednesday May 29th by email please.