{"ID":2842908,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.09447","arxiv_id":"2511.09447","title":"SpaDA: A Spatial Dataflow Architecture Programming Language","abstract":"Spatial dataflow architectures like the Cerebras Wafer-Scale Engine deliver exceptional performance in AI and scientific computing by distributing scratchpad memory across hundreds of thousands of processing elements (PEs). Yet programming these architectures remains difficult: with no shared memory, data movement requires explicit configuration, and asynchronous task management introduces substantial complexity. We present SpaDA, a programming language that offers precise control over data placement, dataflow patterns, and asynchronous operations while abstracting low-level architectural details. We design and implement a compiler targeting Cerebras CSL through multi-level lowering and unique optimization passes. SpaDA functions as a high-level programming interface and an intermediate representation for domain-specific languages (DSLs), demonstrated here with the GT4Py stencil DSL. SpaDA enables concise expression of operations with complex parallel patterns -- including pipelined collective operations, multi-dimensional stencils, and dense linear algebra -- in 14.09x fewer lines than CSL, achieving over 260 TFlop/s across 730,000 PEs on a single device.","short_abstract":"Spatial dataflow architectures like the Cerebras Wafer-Scale Engine deliver exceptional performance in AI and scientific computing by distributing scratchpad memory across hundreds of thousands of processing elements (PEs). Yet programming these architectures remains difficult: with no shared memory, data movement requ...","url_abs":"https://arxiv.org/abs/2511.09447","url_pdf":"https://arxiv.org/pdf/2511.09447v2","authors":"[\"Lukas Gianinazzi\",\"Tal Ben-Nun\",\"Torsten Hoefler\"]","published":"2025-11-12T16:03:52Z","proceeding":"cs.DC","tasks":"[\"cs.DC\",\"cs.PL\"]","methods":"[]","has_code":false}