Skip to main content
  1. Home
  2. Computing

The future of fast PC graphics? Connecting directly to SSDs

Monica J. White
By

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!
Privacy Policy

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.

Editors' Recommendations

Topics
Monica J. White
Monica J. White
Computing Writer
Monica is a UK-based freelance writer and self-proclaimed geek. A firm believer in the "PC building is just like expensive…
GPU prices are back on the rise again
RTX 4060 Ti sitting next to the RTX 4070.

We haven't had to worry about the prices of some of the best graphics cards for quite some time. With most GPUs sold around their recommended retail price, there are plenty of options for PC builders in need of a new graphics card. However, a new report indicates that we might see an increase in GPU prices, especially on the cards made by Nvidia's add-in board partners (AIBs). Is it time to start worrying about another GPU shortage? Not quite, but it might be better to shop now before it gets worse.

The grim news comes from IT Home, a Chinese tech publication that cites anonymous "industry sources" as it predicts that Nvidia's AIBs are about to raise their prices by up to 10% on average -- and this won't be limited to high-end GPUs along the lines of the RTX 4090. In fact, IT Home reports that the RTX 4070 Super has already received a price increase of about 100 yuan, which equals roughly $14 at the time of this writing. This is a subtle price increase given that the GPU costs $550 to $600, but according to the report, it might just be the beginning.

Read more
The sad reality of AMD’s next-gen GPUs comes into view
The AMD RX 7900 graphics card on a pink background.

For months now, various leakers agreed on one thing -- AMD is tapping out of the high-end GPU race in this generation, leaving Nvidia to focus on making the best graphics cards with no competitor. Today's new finding may confirm that theory, as the first RDNA 4 GPU to make an official appearance is one that has been speculated about for months: Navi48.

Following the typical naming convention for AMD, the flagship in the RDNA 4 generation should have been called Navi41 -- and it very well might have been, but according to various sources, that GPU will not be making an appearance in this generation. Hence, the flagship is now said to be the Navi48, and the latest finding shared by Kepler_L2 on X tells us that might indeed be the case.

Read more
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