CARINA: An Efficient CXL-Oriented Embedding Serving System for Recommendation Models
Summary: CARINA optimizes ERM serving on CXL by using heterogeneous memory: hot embeddings on DRAM and NUMA-aware placement of tables. Bandwidth-aware decomposition and scheduling prevent CXL saturation, yielding 5x throughput and 4x latency gains on real devices. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Peiqi Yin
- 2. Qihui Zhou
- 3. Xiao Yan
- 4. Chao Wang
- 5. Eric Lo
- 6. Changji Li
- 7. Lan Lu
- 8. Hua Fan
- 9. Wenchao Zhou
- 10. Ming-Chang Yang
- 11. James Cheng
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 7 of 7 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,320 | High-Throughput Vector Similarity Search in Knowledge Graphs | 2023 | SIGMOD | 9.0366225e-05 |
| 2,688 | Accelerating Recommendation System Training by Leveraging Popular Choices | 2022 | VLDB | 8.2991144e-05 |
| 6,998 | PetPS: Supporting Huge Embedding Models with Persistent Memory | 2023 | VLDB | 4.8676312e-05 |
| 6,999 | WiscSort: External Sorting For Byte-Addressable Storage | 2023 | VLDB | 4.8676312e-05 |
| 9,094 | FEC: Efficient Deep Recommendation Model Training with Flexible Embedding Communication | 2023 | SIGMOD | 4.3980444e-05 |
| 10,974 | GE2: A General and Efficient Knowledge Graph Embedding Learning System | 2024 | SIGMOD | 4.1945683e-05 |
| 10,979 | Atom: An Efficient Query Serving System for Embedding-based Knowledge Graph Reasoning with Operator-level Batching | 2024 | SIGMOD | 4.1945683e-05 |
Previous
Page 1 / 1
Next