使用Accelerate库在多GPU上进行LLM推理

人工智能
大型语言模型(llm)已经彻底改变了自然语言处理领域。随着这些模型在规模和复杂性上的增长,推理的计算需求也显著增加。为了应对这一挑战利用多个gpu变得至关重要。

大型语言模型(llm)已经彻底改变了自然语言处理领域。随着这些模型在规模和复杂性上的增长,推理的计算需求也显著增加。为了应对这一挑战利用多个gpu变得至关重要。

所以本文将在多个gpu上并行执行推理,主要包括:Accelerate库介绍,简单的方法与工作代码示例和使用多个gpu的性能基准测试。

本文将使用多个3090将llama2-7b的推理扩展在多个GPU上

基本示例

我们首先介绍一个简单的示例来演示使用Accelerate进行多gpu“消息传递”。

from accelerate import Accelerator
 from accelerate.utils import gather_object
 
 accelerator = Accelerator()
 
 # each GPU creates a string
 message=[ f"Hello this is GPU {accelerator.process_index}" ] 
 
 # collect the messages from all GPUs
 messages=gather_object(message)
 
 # output the messages only on the main process with accelerator.print() 
 accelerator.print(messages)

输出如下:

['Hello this is GPU 0', 
  'Hello this is GPU 1', 
  'Hello this is GPU 2', 
  'Hello this is GPU 3', 
  'Hello this is GPU 4']

多GPU推理

下面是一个简单的、非批处理的推理方法。代码很简单,因为Accelerate库已经帮我们做了很多工作,我们直接使用就可以:

from accelerate import Accelerator
 from accelerate.utils import gather_object
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from statistics import mean
 import torch, time, json
 
 accelerator = Accelerator()
 
 # 10*10 Prompts. Source: https://www.penguin.co.uk/articles/2022/04/best-first-lines-in-books
 prompts_all=[
    "The King is dead. Long live the Queen.",
    "Once there were four children whose names were Peter, Susan, Edmund, and Lucy.",
    "The story so far: in the beginning, the universe was created.",
    "It was a bright cold day in April, and the clocks were striking thirteen.",
    "It is a truth universally acknowledged, that a single man in possession of a good fortune, must be in want of a wife.",
    "The sweat wis lashing oafay Sick Boy; he wis trembling.",
    "124 was spiteful. Full of Baby's venom.",
    "As Gregor Samsa awoke one morning from uneasy dreams he found himself transformed in his bed into a gigantic insect.",
    "I write this sitting in the kitchen sink.",
    "We were somewhere around Barstow on the edge of the desert when the drugs began to take hold.",
 ] * 10
 
 # load a base model and tokenizer
 model_path="models/llama2-7b"
 model = AutoModelForCausalLM.from_pretrained(
    model_path,    
    device_map={"": accelerator.process_index},
    torch_dtype=torch.bfloat16,
 )
 tokenizer = AutoTokenizer.from_pretrained(model_path)   
 
 # sync GPUs and start the timer
 accelerator.wait_for_everyone()
 start=time.time()
 
 # divide the prompt list onto the available GPUs 
 with accelerator.split_between_processes(prompts_all) as prompts:
    # store output of generations in dict
    results=dict(outputs=[], num_tokens=0)
 
    # have each GPU do inference, prompt by prompt
    for prompt in prompts:
        prompt_tokenized=tokenizer(prompt, return_tensors="pt").to("cuda")
        output_tokenized = model.generate(**prompt_tokenized, max_new_tokens=100)[0]
 
        # remove prompt from output 
        output_tokenized=output_tokenized[len(prompt_tokenized["input_ids"][0]):]
 
        # store outputs and number of tokens in result{}
        results["outputs"].append( tokenizer.decode(output_tokenized) )
        results["num_tokens"] += len(output_tokenized)
 
    results=[ results ] # transform to list, otherwise gather_object() will not collect correctly
 
 # collect results from all the GPUs
 results_gathered=gather_object(results)
 
 if accelerator.is_main_process:
    timediff=time.time()-start
    num_tokens=sum([r["num_tokens"] for r in results_gathered ])
 
    print(f"tokens/sec: {num_tokens//timediff}, time {timediff}, total tokens {num_tokens}, total prompts {len(prompts_all)}")

使用多个gpu会导致一些通信开销:性能在4个gpu时呈线性增长,然后在这种特定设置中趋于稳定。当然这里的性能取决于许多参数,如模型大小和量化、提示长度、生成的令牌数量和采样策略,所以我们只讨论一般的情况

1 GPU: 44个token /秒,时间:225.5s

2 gpu: 88个token /秒,时间:112.9s

3 gpu: 128个token /秒,时间:77.6s

4 gpu: 137个token /秒,时间:72.7s

5 gpu: 119个token /秒,时间:83.8s

在多GPU上进行批处理

现实世界中,我们可以使用批处理推理来加快速度。这会减少GPU之间的通讯,加快推理速度。我们只需要增加prepare_prompts函数将一批数据而不是单条数据输入到模型即可:

from accelerate import Accelerator
 from accelerate.utils import gather_object
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from statistics import mean
 import torch, time, json
 
 accelerator = Accelerator()
 
 def write_pretty_json(file_path, data):
    import json
    with open(file_path, "w") as write_file:
        json.dump(data, write_file, indent=4)
 
 # 10*10 Prompts. Source: https://www.penguin.co.uk/articles/2022/04/best-first-lines-in-books
 prompts_all=[
    "The King is dead. Long live the Queen.",
    "Once there were four children whose names were Peter, Susan, Edmund, and Lucy.",
    "The story so far: in the beginning, the universe was created.",
    "It was a bright cold day in April, and the clocks were striking thirteen.",
    "It is a truth universally acknowledged, that a single man in possession of a good fortune, must be in want of a wife.",
    "The sweat wis lashing oafay Sick Boy; he wis trembling.",
    "124 was spiteful. Full of Baby's venom.",
    "As Gregor Samsa awoke one morning from uneasy dreams he found himself transformed in his bed into a gigantic insect.",
    "I write this sitting in the kitchen sink.",
    "We were somewhere around Barstow on the edge of the desert when the drugs began to take hold.",
 ] * 10
 
 # load a base model and tokenizer
 model_path="models/llama2-7b"
 model = AutoModelForCausalLM.from_pretrained(
    model_path,    
    device_map={"": accelerator.process_index},
    torch_dtype=torch.bfloat16,
 )
 tokenizer = AutoTokenizer.from_pretrained(model_path)   
 tokenizer.pad_token = tokenizer.eos_token
 
 # batch, left pad (for inference), and tokenize
 def prepare_prompts(prompts, tokenizer, batch_size=16):
    batches=[prompts[i:i + batch_size] for i in range(0, len(prompts), batch_size)]  
    batches_tok=[]
    tokenizer.padding_side="left"     
    for prompt_batch in batches:
        batches_tok.append(
            tokenizer(
                prompt_batch, 
                return_tensors="pt", 
                padding='longest', 
                truncatinotallow=False, 
                pad_to_multiple_of=8,
                add_special_tokens=False).to("cuda") 
            )
    tokenizer.padding_side="right"
    return batches_tok
 
 # sync GPUs and start the timer
 accelerator.wait_for_everyone()    
 start=time.time()
 
 # divide the prompt list onto the available GPUs 
 with accelerator.split_between_processes(prompts_all) as prompts:
    results=dict(outputs=[], num_tokens=0)
 
    # have each GPU do inference in batches
    prompt_batches=prepare_prompts(prompts, tokenizer, batch_size=16)
 
    for prompts_tokenized in prompt_batches:
        outputs_tokenized=model.generate(**prompts_tokenized, max_new_tokens=100)
 
        # remove prompt from gen. tokens
        outputs_tokenized=[ tok_out[len(tok_in):] 
            for tok_in, tok_out in zip(prompts_tokenized["input_ids"], outputs_tokenized) ] 
 
        # count and decode gen. tokens 
        num_tokens=sum([ len(t) for t in outputs_tokenized ])
        outputs=tokenizer.batch_decode(outputs_tokenized)
 
        # store in results{} to be gathered by accelerate
        results["outputs"].extend(outputs)
        results["num_tokens"] += num_tokens
 
    results=[ results ] # transform to list, otherwise gather_object() will not collect correctly
 
 # collect results from all the GPUs
 results_gathered=gather_object(results)
 
 if accelerator.is_main_process:
    timediff=time.time()-start
    num_tokens=sum([r["num_tokens"] for r in results_gathered ])
 
    print(f"tokens/sec: {num_tokens//timediff}, time elapsed: {timediff}, num_tokens {num_tokens}")

可以看到批处理会大大加快速度。

1 GPU: 520 token /sec,时间:19.2s

2 gpu: 900 token /sec,时间:11.1s

3 gpu: 1205个token /秒,时间:8.2s

4 gpu: 1655 token /sec,时间:6.0s

5 gpu: 1658 token /sec,时间:6.0s

总结

截止到本文为止,llama.cpp,ctransformer还不支持多GPU推理,好像llama.cpp在6月有个多GPU的merge,但是我没看到官方更新,所以这里暂时确定不支持多GPU。如果有小伙伴确认可以支持多GPU请留言。

huggingface的Accelerate包则为我们使用多GPU提供了一个很方便的选择,使用多个GPU推理可以显着提高性能,但gpu之间通信的开销随着gpu数量的增加而显著增加。

责任编辑:华轩 来源: DeepHub IMBA
相关推荐

2024-03-25 14:22:07

大型语言模型GaLore

2020-03-07 18:51:11

EclipseFedoraPHP

2009-01-06 10:04:44

CygwinGCCGUI

2024-02-04 00:00:00

Triton格式TensorRT

2022-02-09 15:29:35

Java组件编程语言

2010-02-24 15:19:38

ibmdwLinux

2010-12-09 09:12:28

2024-10-16 21:49:24

2009-04-14 18:50:55

Nehalem惠普intel

2020-02-18 09:45:44

云计算云平台IT

2023-06-20 08:00:00

2023-09-01 15:22:49

人工智能数据

2024-01-11 16:24:12

人工智能RAG

2023-06-21 13:44:57

模型AI

2024-08-28 13:34:13

2024-07-26 08:59:33

2011-10-08 11:05:04

GPUMATLAB

2024-08-13 08:23:43

LLamaSharpLLM推理库

2023-12-22 09:32:13

引擎模型

2023-11-27 13:19:54

模型训练
点赞
收藏

51CTO技术栈公众号