Artificial intelligence (AI) has hit the headlines and the datacentres, but with it comes a range of performance and operating considerations that impact storage as much as any other IT discipline.
In this review, we look at the key demands of AI processing on data storage, the type of storage AI requires, and the suitability of cloud storage for AI workloads.
We drill down into the data needs of AI and storage, such as the demands of high-dimension vector data and checkpointing during AI training, plus the compliance considerations that use of AI brings with it.
We also look at the responses of storage suppliers to the rapid rise of AI use cases in the datacentre, in terms of link-ups with leading players like Nvidia, as well as in their storage offer aimed at AI workloads.
1. Storage technology explained: AI and data storage
In this guide, we examine the data storage needs of artificial intelligence, the demands it places on data storage, the suitability of cloud and object storage for AI, and key AI storage products.
2. Storage technology explained: Vector databases at the core of AI
We look at the use of vector data in AI and how vector databases work, plus vector embedding, the challenges for storage of vector data and the key suppliers of vector database products.
3. Interview: Nvidia on AI workloads and their impacts on data storage
We talk to Charlie Boyle of Nvidia about data challenges in artificial intelligence, key practical tips for AI projects, and demands on storage of training, inferencing, RAG and checkpointing.
4. GTC 2024: Storage suppliers queue up to ride the Nvidia AI wave
Storage supplier announcements at Nvdia conference centre on infrastructure integration, tackling the GPU I/O bottleneck and AI hallucinations by running Nvidia NeMo and NIM microservices.
5. Pure CEO: AI needs write speed and storage in place
We spoke to Pure Storage CEO Charlie Giancarlo about why write speed is key for artificial intelligence workloads, accessible storage for AI data, and his prediction of the death of spinning disk.
6. Podcast: Storage for AI needs scale, hybrid cloud and multiple integrations
We talk to NetApp’s Grant Caley about AI and data storage, the need for scale, performance and hybrid cloud, and to move, copy and clone data for wrangling for inference runs.
7. Vast targets AI checkpointing write performance with distributed RAID
AI checkpointing operations targeted by Vast Data as it touts QLC-based storage for AI workloads.
8. Podcast: What is the impact of AI on storage and compliance?
Start looking at artificial intelligence compliance. That’s the advice of Mathieu Gorge of Vigitrust, who says AI governance is still immature, but firms should recognise the limits and still act.
9. AI firm saves a million in shift to Pure FlashBlade shared storage
AI consultancy Crater Labs spent vast amounts of time managing server-attached drives to ensure GPUs were saturated. A shift to all-flash Pure Storage slashed that to almost zero.
10. Google launches Parallelstore file storage at cloud AI training
Originally driven by Intel’s now-defunct Optane storage class memory, Parallelstore offers massive parallel file storage targeted at artificial intelligence training use cases on Google Cloud.
