As Mixture-of-Experts (MoE) models continue to scale, serving them increasingly relies on expert parallelism (EP) across a growing number of devices. Yet skewed expert workloads create imbalance across computation, communication, and memory, making load balancing a central optimization objective in distributed MoE serv...
Chang Chen, Andrew Yang, Tian-Cheng Chen et al.· 0 citations
This work presents RR-EVICT, a fine-grained prefix-cache eviction strategy that distributes reclamation across idle agent trajectories, and implements RR-EVICT in SGLang and evaluates conversational and coding-agent workloads under colocated and prefill-decode-disaggregated serving.
Zai-Feng Pan, Chris Wu, Zheng-Ding Hu et al.· 0 citations
The experiments show that token-wise replay in AgentReplay effectively eliminates workload variation that greedy decoding and length-wise replay cannot avoid, enabling fairer performance comparisons across serving configurations.
Zai-Feng Pan, Michael Wang, Chris Wu et al.· 1 citation
This work proposes a series of optimizations for these two kernels, including computation-transfer pipelining, load balancing, and memory access fusion, achieving 1.97 × to 2.16 × proof generation speedup over a state-of-the-art open source GPU acceleration library.
Xinwei Qiang, Liukun Yu, Xiyu Wang et al.· IEEE International Symposium...· 0 citations
Block drafters propose several tokens in one forward pass, before earlier target tokens are realised. Their rejection mixes two losses: missing within-block path information and imperfect modelling of observable information. Accepted length cannot distinguish them. We separate the two with an information floor, the min...
Xinwei Qiang, Xiang Fang, Changli Chen et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
