- Kioxia GP Sequence SSD gives GPUs with quicker reminiscence entry past HBM limits
- Storage Class Reminiscence bridges the efficiency hole between DRAM and standard NAND flash storage
- XL-FLASH prioritizes low latency and tens of millions of random IOPS over sequential pace
Kioxia has launched a brand new sort of solid-state drive designed to operate as a direct reminiscence enlargement for GPUs.
The brand new Kioxia GP Sequence, introduced at Nvidia GTC 2026, shouldn’t be a alternative for current storage, however slightly a further tier within the reminiscence hierarchy.
Its major function is to offer a bigger pool of fast-accessible information for GPUs, successfully performing as an overflow for the costly and capacity-limited Excessive Bandwidth Reminiscence.
Article continues under
You might like
Reminiscence-hungry AI fashions drive the change
The drive leverages Storage Class Reminiscence (SCM), a class of expertise that sits within the efficiency hole between conventional NAND flash and system DRAM.
This idea was popularized years in the past by Intel’s now-discontinued Optane expertise, which aimed to bridge the identical divide, however failed.
Kioxia’s model, branded as XL-FLASH, prioritizes low latency and excessive enter/output operations per second over uncooked sequential throughput, permitting finer-grained information entry at simply 512bytes.
This growth is a direct response to a elementary drawback in present AI infrastructure, GPU reminiscence is just not sufficiently big for the fashions it’s requested to run.
As AI fashions scale towards trillions of parameters and context home windows broaden to tens of millions of tokens, the demand for reminiscence to retailer objects just like the KV cache has outpaced the bodily limits of HBM.
Nvidia’s Storage-Subsequent initiative, which this SSD helps, was created to deal with this precise bottleneck by encouraging storage distributors to construct drives that GPUs can hyperlink to immediately.
“Kioxia totally helps the NVIDIA Storage-Subsequent initiative and can ship purpose-built SSDs to successfully handle the necessity for GPU-accessible reminiscence,” stated Makoto Hamada, Senior Director of the SSD Division, Kioxia Company.
What to learn subsequent
Whereas the GP Sequence goals for tens of millions of IOPS to feed information to GPUs, the broader business is chasing much more bold efficiency targets.
The idea of reaching 100 million IOPS is a identified business pipedream, one which will require inventing totally new lessons of reminiscence.
Different corporations are additionally focusing on the area of interest market that Optane left behind.
For instance, the InnoGrit N3X SSD makes use of Kioxia’s XL-Flash in SLC mode to ship excessive endurance.
This drive can reportedly face up to as much as 50 full drive writes per day for 5 consecutive years.
Kioxia itself has beforehand signaled its intent to push towards even increased efficiency figures.
The corporate is aiming for 10 million IOPS utilizing SLC NAND, which means that the GP Sequence is only one step in an extended race to remove storage latency for processors.
It stays to be seen whether or not the Kioxia GP Sequence SSD will succeed the place earlier SCM makes an attempt stumbled.
Comply with TechRadar on Google Information and add us as a most well-liked supply to get our skilled information, evaluations, and opinion in your feeds. Be sure that to click on the Comply with button!
And naturally it’s also possible to comply with TechRadar on TikTok for information, evaluations, unboxings in video kind, and get common updates from us on WhatsApp too.
