Making large AI models cheaper, faster and more accessible
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
Jianghai cf579ff46d
[Inference] Dynamic Batching Inference, online and offline (#4953)
1 year ago
..
dynamic_batching [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago
pipeline [Pipeline inference] Combine kvcache with pipeline inference (#4938) 1 year ago
quant [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago
tensor_parallel [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago
README.md
__init__.py [Pipeline inference] Combine kvcache with pipeline inference (#4938) 1 year ago
async_engine.py [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago
async_manager.py [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago
manager.py [Inference] Dynamic Batching Inference, online and offline (#4953) 1 year ago

README.md

🚀 Colossal-Inference

Table of contents

Introduction

Colossal Inference is a module that contains colossal-ai designed inference framework, featuring high performance, steady and easy usability. Colossal Inference incorporated the advantages of the latest open-source inference systems, including LightLLM, TGI, vLLM, FasterTransformer and flash attention. while combining the design of Colossal AI, especially Shardformer, to reduce the learning curve for users.

Design

Colossal Inference is composed of two main components:

  1. High performance kernels and ops: which are inspired from existing libraries and modified correspondingly.
  2. Efficient memory management mechanism:which includes the key-value cache manager, allowing for zero memory waste during inference.
    1. cache manager: serves as a memory manager to help manage the key-value cache, it integrates functions such as memory allocation, indexing and release.
    2. batch_infer_info: holds all essential elements of a batch inference, which is updated every batch.
  3. High-level inference engine combined with Shardformer: it allows our inference framework to easily invoke and utilize various parallel methods.
    1. engine.TPInferEngine: it is a high level interface that integrates with shardformer, especially for multi-card (tensor parallel) inference:
    2. modeling.llama.LlamaInferenceForwards: contains the forward methods for llama inference. (in this case : llama)
    3. policies.llama.LlamaModelInferPolicy : contains the policies for llama models, which is used to call shardformer and segmentate the model forward in tensor parallelism way.

Pipeline of inference:

In this section we discuss how the colossal inference works and integrates with the Shardformer . The details can be found in our codes.

Colossal-Inference

Roadmap of our implementation

  • Design cache manager and batch infer state
  • Design TpInference engine to integrates with Shardformer
  • Register corresponding high-performance kernel and ops
  • Design policies and forwards (e.g. Llama and Bloom)
    • policy
    • context forward
    • token forward
  • Replace the kernels with faster-transformer in token-forward stage
  • Support all models
    • Llama
    • Bloom
    • Chatglm2
  • Benchmarking for all models

Get started

Installation

pip install -e .

Requirements

dependencies

pytorch= 1.13.1 (gpu)
cuda>= 11.6
transformers= 4.30.2
triton==2.0.0.dev20221202
# for install vllm, please use this branch to install https://github.com/tiandiao123/vllm/tree/setup_branch
vllm
# for install flash-attention, please use commit hash: 67ae6fd74b4bc99c36b2ce524cf139c35663793c
flash-attention

# install lightllm since we depend on lightllm triton kernels
git clone https://github.com/ModelTC/lightllm 
git checkout 28c1267cfca536b7b4f28e921e03de735b003039
cd lightllm
pip3 install -e .

Docker

You can use docker run to use docker container to set-up environment

# env: python==3.8, cuda 11.6, pytorch == 1.13.1 triton==2.0.0.dev20221202, vllm kernels support, flash-attention-2 kernels support
docker pull hpcaitech/colossalai-inference:v2
docker run -it --gpus all --name ANY_NAME -v $PWD:/workspace -w /workspace hpcaitech/colossalai-inference:v2 /bin/bash

# enter into docker container
cd /path/to/CollossalAI
pip install -e .

# install lightllm
git clone https://github.com/ModelTC/lightllm 
git checkout 28c1267cfca536b7b4f28e921e03de735b003039
cd lightllm
pip3 install -e .


Dive into fast-inference!

example files are in

cd colossalai.examples
python xx

Performance

environment:

We conducted multiple benchmark tests to evaluate the performance. We compared the inference latency and throughputs between colossal-inference and original hugging-face torch fp16.

For various models, experiments were conducted using multiple batch sizes under the consistent model configuration of 7 billion(7b) parameters, 1024 input length, and 128 output length. The obtained results are as follows (due to time constraints, the evaluation has currently been performed solely on the A100 single GPU performance; multi-GPU performance will be addressed in the future):

Single GPU Performance:

Currently the stats below are calculated based on A100 (single GPU), and we calculate token latency based on average values of context-forward and decoding forward process, which means we combine both of processes to calculate token generation times. We are actively developing new features and methods to further optimize the performance of LLM models. Please stay tuned.

Llama

batch_size 8 16 32
hugging-face torch fp16 199.12 246.56 278.4
colossal-inference 326.4 582.72 816.64

llama

Bloom

batch_size 8 16 32
hugging-face torch fp16 189.68 226.66 249.61
colossal-inference 323.28 538.52 611.64

bloom

The results of more models are coming soon!