- WEKApod 3 reaches exabyte capacity through software along with dense Micron storage
- NeuralMesh 6 turns 441.5 PB of flash into claimed exabyte-scale storage capacity
- WEKApod 3 relies on compression to exceed the limits of raw stock hardware
WEKA has launched the WEKApod 3, a storage appliance line built specifically to run its new NeuralMesh 6 software at rack-scale density.
The flagship Prime Max configuration claims 1.1 exabytes of effective capacity within a single 56U rack, placing it as the first system to cross this threshold.
Using Micron’s 245.76 TB 6600 ION SSDs at around 1,800 drives per rack provides only 441.5 PB of raw hardware capacity, well under half an exabyte.
Latest videos fromTechRadar
Software tiering closes the capability gap
This means that the advertised exabyte-scale capacity depends heavily on data reduction technologies rather than installed flash alone.
This configuration packs 70 NVMe drives into a compact two-rack, two-node chassis for maximum density.
Fingerprinting, similarity hashing, deduplication and compression run by default across every deployment, with write overhead kept below 5%.
WEKA claims up to 6x capacity savings specifically on AI training data, backed by a contractual guarantee covering both reduction ratio and performance.
This is where NeuralMesh 6 and WEKApod 3 really work as one integrated product rather than separate messages.
AlloyFlash tiering automatically routes latency-sensitive operations to TLC flash while sending bulk capacity data to QLC drives, which run about 30 to 40% cheaper per terabytes.
Per-rack throughput reaches 10.2 TB/s with 210 million IOPS, numbers WEKA says beat the next best publicly available alternative by 114% in throughput density.
Jeremy Werner, senior vice president at Micron’s Core Data Center Business Unit, said the architecture paired with Micron’s 245TB SSDs delivers 15.8 petabytes within a single 2U footprint.
Steve McDowell, principal analyst at NAND Research, argued that production-scale inference workloads require different metrics than training, pointing to tokens per stand and price per conclusion, as the numbers buyers now have to track.
Limitations are forcing storage vendors to completely rethink hardware
WEKA points to several structural pressures behind this hardware approach, starting with a documented decline in US data center construction by 2025.
Grid connection queues in major markets now stretch four to seven years before new capacity comes online.
Morgan Stanley has predicted a 49-gigawatt power outage across the U.S. extending through 2028, exacerbating existing NAND supply constraints.
WEKA claims that inefficient storage directly competes with GPUs for the same limited rack space and power budget.
By managing its own hardware supply chain rather than relying on third-party OEM channels, the company says it can offer more predictable pricing and delivery timelines.
NeuralMesh 6 will hit shelves in the second half of 2026, with existing customers upgrading at no additional cost.
The WEKApod Nitro, Prime and Prime Max can already be ordered through the WEKA distributor network, with deliveries starting in autumn 2026.
Via stock review
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
