TPLA splits the latent KV cache across tensor-parallel GPUs while keeping every head's full view, yielding 1.79x and 1.93x decode speedups on DeepSeek-V3 and Kimi-K2 at 32K context with modest accuracy loss.
An Efficient 2D Method for Training Super-Large Deep Learning Models
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Huge neural network models have shown unprecedented performance in real-world applications. However, due to memory constraints, model parallelism must be utilized to host large models that would otherwise not fit into the memory of a single device. Previous methods like Megatron partition the parameters of the entire model among multiple devices, while each device has to accommodate the redundant activations in forward and backward pass. In this work, we propose Optimus, a highly efficient and scalable 2D-partition paradigm of model parallelism that would facilitate the training of infinitely large language models. In Optimus, activations are partitioned and distributed among devices, further reducing redundancy. In terms of isoefficiency, Optimus significantly outperforms Megatron. On 64 GPUs of TACC Frontera, Optimus achieves 1.48X speedup for training, 1.78X speedup for inference, and 8X increase in maximum batch size over Megatron. Optimus surpasses Megatron in scaling efficiency by a great margin. The code is available at https://github.com/xuqifan897/Optimus.
citation-role summary
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
TPLA: Tensor Parallel Latent Attention for Efficient Disaggregated Prefill and Decode Inference
TPLA splits the latent KV cache across tensor-parallel GPUs while keeping every head's full view, yielding 1.79x and 1.93x decode speedups on DeepSeek-V3 and Kimi-K2 at 32K context with modest accuracy loss.