SnipSnap: A Joint Compression Format and Dataflow Co-Optimization Framework for Efficient Sparse LLM Accelerator Design
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910050030714880 |
|---|---|
| author | Wu, Junyi Fang, Chao Wang, Zhongfeng |
| author_facet | Wu, Junyi Fang, Chao Wang, Zhongfeng |
| contents | The growing scale of large language models (LLMs) has intensified demands on computation and memory, making efficient inference a key challenge. While sparsity can reduce these costs, existing design space exploration (DSE) frameworks often overlook compression formats, a key factor for leveraging sparsity on accelerators. This paper proposes SnipSnap, a joint compression format and dataflow co-optimization framework for efficient sparse LLM accelerator design. SnipSnap introduces: (1) a hierarchical compression format encoding to expand the design space; (2) an adaptive compression engine for selecting formats under diverse sparsity; and (3) a progressive co-search workflow that jointly optimizes dataflow and compression formats. SnipSnap achieves 18.24% average memory energy savings via format optimization, along with 2248.3$\times$ and 21.0$\times$ speedups over Sparseloop and DiMO-Sparse frameworks, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_17072 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SnipSnap: A Joint Compression Format and Dataflow Co-Optimization Framework for Efficient Sparse LLM Accelerator Design Wu, Junyi Fang, Chao Wang, Zhongfeng Hardware Architecture The growing scale of large language models (LLMs) has intensified demands on computation and memory, making efficient inference a key challenge. While sparsity can reduce these costs, existing design space exploration (DSE) frameworks often overlook compression formats, a key factor for leveraging sparsity on accelerators. This paper proposes SnipSnap, a joint compression format and dataflow co-optimization framework for efficient sparse LLM accelerator design. SnipSnap introduces: (1) a hierarchical compression format encoding to expand the design space; (2) an adaptive compression engine for selecting formats under diverse sparsity; and (3) a progressive co-search workflow that jointly optimizes dataflow and compression formats. SnipSnap achieves 18.24% average memory energy savings via format optimization, along with 2248.3$\times$ and 21.0$\times$ speedups over Sparseloop and DiMO-Sparse frameworks, respectively. |
| title | SnipSnap: A Joint Compression Format and Dataflow Co-Optimization Framework for Efficient Sparse LLM Accelerator Design |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2509.17072 |