shards.si
A shard is a partition of a database, dataset, or model's parameters, held on a separate machine so the whole doesn't have to fit on one. It's the mechanism behind sharded databases, and behind every model too large for a single GPU.
Why "shards"
No frontier model runs on one machine. Training splits a model's parameters and optimizer states across thousands of GPUs — sharded, by name, in every major framework: FSDP, DeepSpeed, tensor and data parallelism. Inference does the same across model and pipeline shards to serve anything too large for a single device.
This isn't a phase superintelligence-scale computing grows out of. It's the opposite — as models get larger and workloads spread further across hardware, the amount of computing that happens inside a shard, rather than on a whole machine, only grows. The term describes the substrate itself, not a workaround.
Contact
shards.si is for sale, privately. Contact for registration history and transfer terms.
The portfolio
misalignment.si, plus nine names for the layers of modern infrastructure, each offered privately.