Skip to main content

The future of fast PC graphics? Connecting directly to SSDs

Performance boosts are expected with each new generation of the best graphics cards, but it seems that Nvidia and IBM have their sights set on greater changes.

The companies teamed up to work on Big accelerator Memory (BaM), a technology that involves connecting graphics cards directly to superfast SSDs. This could result in larger GPU memory capacity and faster bandwidth while limiting the involvement of the CPU.

A chart breaks down Nvidia and IBM's BaM technology.
Image source: Arxiv Image used with permission by copyright holder

This type of technology has already been thought of, and worked on, in the past. Microsoft’s DirectStorage application programming interface (API) works in a somewhat similar way, improving data transfers between the GPU and the SSD. However, this relies on external software, only applies to games, and only works on Windows. Nvidia and IBM researchers are working together on a solution that removes the need for a proprietary API while still connecting GPUs to SSDs.

Recommended Videos

The method, amusingly referred to as BaM, was described in a paper written by the team that designed it. Connecting a GPU directly to an SSD would provide a performance boost that could prove to be viable, especially for resource-heavy tasks such as machine learning. As such, it would mostly be used in professional high-performance computing (HPC) scenarios.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

The technology that is currently available for processing such heavy workloads requires the graphics card to rely on large amounts of special-purpose memory, such as HBM2, or to be provided with efficient access to SSD storage. Considering that datasets are only growing in size, it’s important to optimize the connection between the GPU and storage in order to allow for efficient data transfers. This is where BaM comes in.

“BaM mitigates the I/O traffic amplification by enabling the GPU threads to read or write small amounts of data on-demand, as determined by the compute,” said the researchers in their paper, first cited by The Register. “The goal of BaM is to extend GPU memory capacity and enhance the effective storage access bandwidth while providing high-level abstractions for the GPU threads to easily make on-demand, fine-grain access to massive data structures in the extended memory hierarchy.”

An Nvidia GPU core sits on a table.
Niels Broekhuijsen / Digital Trends

For many people who don’t work directly with this subject, the details may seem complicated, but the gist of it is that Nvidia wants to rely less on the processor and connect directly to the source of the data. This would both make the process more efficient and free up the CPU, making the graphics card much more self-sufficient. The researchers claim that this design would be able to compete with DRAM-based solutions while remaining cheaper to implement.

Although Nvidia and IBM are undoubtedly breaking new ground with their BaM technology, AMD worked in this area first: In 2016, it unveiled the Radeon Pro SSG, a workstation GPU with integrated M.2 SSDs. However, the Radeon Pro SSG was intended to be strictly a graphics solution, and Nvidia is taking it a few steps further, aiming to deal with complex and heavy compute workloads.

The team working on BaM plans to release the details of their software and hardware optimization as open source, allowing others to build on their findings. There is no mention as to when, if ever, BaM might find itself implemented in future Nvidia products.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
Everything you need to know about buying a GPU in 2024
RTX 4090.

The graphics card, also known as the GPU, is arguably one of the most exciting components in any PC build. Alongside the processor, your graphics card often has the greatest impact on the overall performance of your PC. That makes it a pretty high-stakes purchase, especially if you consider that GPUs can get pretty expensive.

The GPU market has a lot to offer, and that's regardless of your needs and your budget. Whether you're aiming for something super cheap to support some light browsing or a behemoth to handle the most GPU-intensive games, you have lots of options. In this guide, we'll show you what to look out for so that you can pick the best GPU that fits your budget and needs.
Nvidia, AMD, or Intel?
Consumer graphics cards are generally split into two categories -- integrated and discrete graphics. Since you're here, you're most likely looking for a discrete (or dedicated) GPU, and that's what we're going to focus on in this article.

Read more
I’m jealous — someone scored an RTX 4070 Super for $49 on Amazon
The Nvidia logo on the RTX 4070 Super.

The RTX 4070 Super is one of the best graphics cards you can buy, and it has a price to match, with most models going for $600 on Amazon. However, one lucky Redditor scored a model for just $49. No, it wasn't some steep sale, and most people would see a price like that and assume it's a scam. But sure enough, two days after finding a Gigabyte RTX 4070 Super Gaming OC for $48.94 on Amazon, the card showed up in the mail.

You should absolutely assume that a price that's too good to be true is a scam on Amazon. We've seen just this year how scammers can infiltrate the third-party listings on Amazon to sell fake graphics cards, but this listing for the RTX 4070 Super was different. It was sold and shipped by Amazon Canada and there was only one card in stock, suggesting it was either mismarked or someone seriously needed some extra warehouse space.

Read more
Intel Arc Battlemage finally shows signs of life
The Intel logo on the Arc A770 graphics card.

In the last few weeks, we've been bombarded with reports regarding Nvidia's upcoming top graphics cards, but leakers have all been silent about Intel Arc Battlemage. Today, however, we saw signs of life straight from the source as Intel has included Battlemage graphics cards in its new media drivers.

As spotted by VideoCardz, Intel added BMG -- the shortcut for Battlemage -- to its latest round of drivers on GitHub. This covers an addition to the open-source media driver for the video acceleration API (VAAPI), which includes various codecs for video decoding and encoding. The column labeled as BMG looks pretty bare, though, with zero encoding support so far.

Read more