- SolidM’s new AI -Lab redefines what close SSD -storage can achieve
- The cluster delivers record flow but still lingering over real scalability
- 23.6 Petabytes pressed in 16u challenges how storage is seen
Solidigm has opened its AI Central Lab at the FarmgPU site near its headquarters in Rancho Cordova, California.
The facility is presented as a place to study how storage interacts with artificial intelligence workload using high performance GPUs and close storage arrays.
The company says the laboratory delivers one of the most compact large clusters in the industry, which is intended to replicate conditions found in modern data centers.
Testing the claims of record performance
At the core of the message is the Solidants D7-PS1010 SSDs used to reach 116 GB/s per day. Knude in MLPERF storage benchmarks.
This number is described as a record result, although the value of such a synthetic test for the real world AI operations remain open to the debate.
Benchmarks often highlight top flow, but the actual workload performance can depend strongly on software and system integration.
Lab still offers a platform where storage providers, developers and partners can run experiments under controlled but relevant conditions.
“Our Soliktm AI Central Lab combines today’s most powerful GPUs with leading storage infrastructure to lock new levels of testing and joint innovation for our customers and the developer community,” said AVI Shetty, senior director, AI ecosystems and partnerships, Soliktm.
“These were capabilities that were previously only available to select companies, and Solidigm now enables them while demonstrating the criticism of having storage close to GPU.”
The most conspicuous aspect is perhaps the claim of 23.6 PB storage packed in only 16U rack space.
This was achieved using 192 Solidigm D5-P5336 SSDs, each with 122 TB capacity.
By volume, it can represent one of the closest corporate storage clusters implemented in public.
This raises questions about how such hardware compares to the best HDD settings that continue to offer lower costs per day. Terabytes.
If the benefit requires outweighing cost-effectiveness, the SolidM’s setup could be seen as a step towards defining the largest SSD-based systems in production environments.
Soliktm emphasized collaboration with partners such as Metrum AI, who claim to have cut drama use during recycling-augmented generation with as much as 57% by relieving data to SSDs.
Such claims suggest potential benefits for memory control, although they also highlight the dependence on the AI ​​efficiency of tightly coupled hardware and software tuning.
While the best SSD devices in the laboratory can demonstrate impressive speed and density, the wider market can weigh these gains against practical considerations such as energy consumption, scalability and long-term costs.
Soliktms AI Central Lab positions itself as a space for both innovation and marketing demonstration.
“Driving storage testing is no longer enough. In our AI Central LAB, we can perform AI workloads in the real world and use our advanced telemetry capacities to optimize performance and efficiency systems and gain insight into the storage needs of new work loads,” SHETTY said.
Follow Techradar on Google News and Add us as a preferred source To get our expert news, reviews and meaning in your feeds. Be sure to click the Follow button!
And of course you can too Follow Techradar at Tiktok For news, reviews, unboxings in video form and get regular updates from us at WhatsApp also.
