ByteDance-Seed / StragglerAnalysisLinks
☆40Updated 4 months ago
Alternatives and similar repositories for StragglerAnalysis
Users that are interested in StragglerAnalysis are comparing it to the libraries listed below
Sorting:
- kvcached: Elastic KV cache for dynamic GPU sharing and efficient multi-LLM inference.☆66Updated this week
- Stateful LLM Serving☆81Updated 5 months ago
- A resilient distributed training framework☆94Updated last year
- ☆69Updated last year
- Microsoft Collective Communication Library☆67Updated 9 months ago
- A lightweight design for computation-communication overlap.☆160Updated last week
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆58Updated 9 months ago
- ☆25Updated 2 years ago
- LLM checkpointing for DeepSpeed/Megatron☆19Updated 2 weeks ago
- ☆123Updated 10 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆181Updated 11 months ago
- A framework for generating realistic LLM serving workloads☆54Updated 2 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆128Updated last year
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆56Updated this week
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆39Updated last week
- ☆55Updated 7 months ago
- ☆47Updated 8 months ago
- nnScaler: Compiling DNN models for Parallel Training☆118Updated this week
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆56Updated 2 months ago
- DeeperGEMM: crazy optimized version☆70Updated 3 months ago
- Open-source implementation for "Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow"☆63Updated 9 months ago
- Supplemental materials for The ASPLOS 2025 / EuroSys 2025 Contest on Intra-Operator Parallelism for Distributed Deep Learning☆22Updated 3 months ago
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆53Updated last year
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆47Updated this week
- SOTA Learning-augmented Systems☆36Updated 3 years ago
- ☆116Updated 9 months ago
- ☆75Updated 3 years ago
- Artifact for "Apparate: Rethinking Early Exits to Tame Latency-Throughput Tensions in ML Serving" [SOSP '24]☆25Updated 9 months ago
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆29Updated last year
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆31Updated 2 months ago