提交 0d3d4ae7 编写于 作者: Q qiaolongfei

refine prefetch logic

上级 17b42fc2
develop 2.0.1-rocm-post Ligoml-patch-1 OliverLPH-patch-1 OliverLPH-patch-2 PaddlePM-patch-1 PaddlePM-patch-2 ZHUI-patch-1 add_default_att add_model_benchmark_ci add_some_yaml_config addfile all_new_design_exec ascendrc ascendrelease cherry_undefined_var compile_windows delete_2.0.1-rocm-post delete_add_default_att delete_all_new_design_exec delete_ascendrc delete_compile_windows delete_delete_addfile delete_disable_iterable_dataset_unittest delete_fix_dataloader_memory_leak delete_fix_imperative_dygraph_error delete_fix_retry_ci delete_fix_undefined_var delete_improve_sccache delete_incubate/lite delete_paddle_tiny_install delete_paralleltest delete_prv-disable-more-cache delete_revert-31068-fix_conv3d_windows delete_revert-31562-mean delete_revert-33630-bug-fix delete_revert-34159-add_npu_bce_logical_dev delete_revert-34910-spinlocks_for_allocator delete_revert-35069-revert-34910-spinlocks_for_allocator delete_revert-36057-dev/read_flags_in_ut dingjiaweiww-patch-1 disable_iterable_dataset_unittest dy2static enable_eager_model_test final_state_gen_python_c final_state_intermediate fix-numpy-issue fix_concat_slice fix_dataloader_memory_leak fix_imperative_dygraph_error fix_npu_ci fix_op_flops fix_retry_ci fix_rnn_docs fix_tensor_type fix_undefined_var fixiscan fixiscan1 fixiscan2 fixiscan3 github/fork/123malin/netifaces github/fork/123malin/tdm_abacus github/fork/AshburnLee/dev_unique github/fork/ForFishes/fix_memory_matmul github/fork/ForFishes/rm_fluid github/fork/LielinJiang/move-2.0-api github/fork/LielinJiang/visual-dl-cb github/fork/LiuChiachi/add-transformer-generate-square-subsequent-mask-api github/fork/LiuChiachi/fix-example-code-for-hapi-Model github/fork/LiuChiachi/remove-input-requirment-in-dygraph-Model github/fork/MrChengmo/fix_ps_profiler github/fork/MrChengmo/update_ps_heter github/fork/PWhiddy/patch-1 github/fork/Shixiaowei02/dev/save_load_upgrade github/fork/TCChenlong/fix_hapi github/fork/TCChenlong/fix_inden github/fork/Thunderbrook/xpu_slice github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_2 github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_3 github/fork/XieYunshen/timeout_20S_ut github/fork/ZeyuChen/remove-nltk github/fork/arlesniak/arlesniak/selective__mkldnn_flags github/fork/baiyfbupt/code_doc_mig github/fork/chalsliu/set_timeout github/fork/chen-zhiyu/develop github/fork/chenwhql/ci/try_to_find_test_buffer_shared_memory_reuse_pass_error github/fork/chenwhql/dygraph/remove_scale_loss_and_apply_collective_grads github/fork/chenwhql/saveload/add_get_inference_program github/fork/chenwhql/saveload/remove_save_load_config github/fork/cryoco/pass-compatibility-trt github/fork/danleifeng/isempty_api2.0 github/fork/frankwhzhang/api_transfer github/fork/hbwx24/error_msg/cuda_kernel_error_msg github/fork/heavengate/cherry_yolo_box github/fork/heavengate/update_yolo_box github/fork/iclementine/rnn_fix github/fork/iducn/testestse github/fork/jczaja/prv-25537-fix github/fork/jeff41404/release/1.8 github/fork/jiweibo/api_2.0 github/fork/jiweibo/fix_lite_resnet50_test github/fork/juncaipeng/fix_doc_1 github/fork/lfchener/sample_code github/fork/littletomatodonkey/fix_reg_doc github/fork/liym27/dy2stat_update_assign_to_rc20 github/fork/luotao1/profiler_ut github/fork/mapingshuo/add_wait github/fork/mapingshuo/doc_2.0 github/fork/mapingshuo/zero-0.5 github/fork/miraiwk/dev github/fork/pangyoki/add-Categorical-class-branch github/fork/pangyoki/add-multinomial-op-branch github/fork/pangyoki/fix-test_distritbution-CI github/fork/qjing666/doublegrad github/fork/qjing666/fix_hdfs_download github/fork/sandyhouse/add_gather_etc github/fork/sandyhouse/add_send_recv_alltoall_etc github/fork/sandyhouse/pipeline_exe_run github/fork/seiriosPlus/feature/large_scale_kv_save_delta github/fork/seiriosPlus/fix/paddle_errors_fix github/fork/seiriosPlus/fix/paddle_op_errors github/fork/shangzhizhou/fix_test_activation_op_random_bug github/fork/smallv0221/yxp0924 github/fork/smallv0221/yxp0925 github/fork/swtkiwi/del-matplotlib github/fork/tianshuo78520a/kunlun_test github/fork/tianshuo78520a/update_dockerfile github/fork/wanghaoshuang/bert_fuse github/fork/wanghaoshuang/label_smooth github/fork/wanghuancoder/develop_CUDASynchronize github/fork/wanghuancoder/develop_Layer_doc github/fork/wanghuancoder/develop_ParameterList_doc github/fork/wanghuancoder/develop_Sequential_doc github/fork/wanghuancoder/develop_bilinear_tensor_product github/fork/wanghuancoder/develop_coverage_build_sh github/fork/wanghuancoder/develop_in_dynamic_mode_doc github/fork/wanghuancoder/develop_unique_name_doc github/fork/wangxicoding/fleet_meta_combine github/fork/wawltor/error_message_fix_5 github/fork/willthefrog/remove_l2_norm github/fork/windstamp/momentum_op github/fork/windstamp/mv_op_5 github/fork/windstamp/normal_api github/fork/wojtuss/wojtuss/fusion_gru_quantization github/fork/wojtuss/wojtuss/quantization-with-shift github/fork/wzzju/fix_err_info github/fork/wzzju/pure_fp16 github/fork/xiemoyuan/op_error_message github/fork/xiemoyuan/optimize_error_message github/fork/yaoxuefeng6/fix_doc github/fork/yaoxuefeng6/mod_dataset_v2 github/fork/yongqiangma/lod github/fork/ysh329/fix-clip-by-norm-error github/fork/ysh329/fix-error-clip-by-value github/fork/yukavio/error_info github/fork/zhangting2020/conv_filter_grad github/fork/zhangting2020/is_compile_with_cuda github/fork/zhangting2020/place_doc github/fork/zhangting2020/program github/fork/zhhsplendid/fix_any github/fork/zhhsplendid/refine_api2 github/fork/zhhsplendid/refine_api2_test github/fork/zhhsplendid/refine_api_test_ptb_lm github/fork/zhhsplendid/refine_api_test_resnet github/fork/zhhsplendid/refine_api_test_simnet github/fork/zhiqiu/dev/refine_initializer github/fork/zhiqiu/dev/remove_inplace_argument github/fork/zlsh80826/nvinfer_plugin_var_len_cuda11 improve_sccache incubate/infrt incubate/lite inplace_addto make_flag_adding_easier move_embedding_to_phi move_histogram_to_pten move_sgd_to_phi move_slice_to_pten move_temporal_shift_to_phi move_yolo_box_to_phi npu_fix_alloc numel paddle_tiny_install paralleltest preln_ernie prv-disable-more-cache prv-md-even-more prv-onednn-2.5 pten_tensor_refactor release/0.14.0 release/0.15.0 release/1.0.0 release/1.1 release/1.2 release/1.3 release/1.4 release/1.5 release/1.6 release/1.7 release/1.8 release/2.0 release/2.0-alpha release/2.0-beta release/2.0-rc release/2.0-rc1 release/2.1 release/2.2 release/2.3 release/2.3-fc-ernie-fix release/2.4 release/lite-0.1 revert-24981-add_device_attr_for_regulization revert-26856-strategy_example2 revert-27520-disable_pr revert-31068-fix_conv3d_windows revert-31562-mean revert-32290-develop-hardlabel revert-33037-forci revert-33475-fix_cifar_label_dimension revert-33630-bug-fix revert-34159-add_npu_bce_logical_dev revert-34406-add_copy_from_tensor revert-34910-spinlocks_for_allocator revert-35069-revert-34910-spinlocks_for_allocator revert-36057-dev/read_flags_in_ut revert-36201-refine_fast_threaded_ssa_graph_executor revert-36985-add_license revert-37318-refactor_dygraph_to_eager revert-37926-eager_coreops_500 revert-37956-revert-37727-pylayer_support_tuple revert-38100-mingdong revert-38301-allocation_rearrange_pr revert-38703-numpy_bf16_package_reupload revert-38732-remove_useless_header_in_elementwise_mul_grad revert-38959-Reduce_Grad revert-39143-adjust_empty revert-39227-move_trace_op_to_pten revert-39268-dev/remove_concat_fluid_kernel revert-40170-support_partial_grad revert-41056-revert-40727-move_some_activaion_to_phi revert-41065-revert-40993-mv_ele_floordiv_pow revert-41068-revert-40790-phi_new revert-41944-smaller_inference_api_test revert-42149-do-not-reset-default-stream-for-stream-safe-cuda-allocator revert-43155-fix_ut_tempfile revert-43882-revert-41944-smaller_inference_api_test revert-45808-phi/simplify_size_op revert-46827-deform_comment rocm_dev_0217 support_weight_transpose test_benchmark_ci test_feature_precision_test_c test_model_benchmark test_model_benchmark_ci zhiqiu-patch-1 v2.4.0-rc0 v2.3.2 v2.3.1 v2.3.0 v2.3.0-rc0 v2.2.2 v2.2.1 v2.2.0 v2.2.0-rc0 v2.2.0-bak0 v2.1.3 v2.1.2 v2.1.1 v2.1.0 v2.1.0-rc0 v2.0.2 v2.0.1 v2.0.0 v2.0.0-rc1 v2.0.0-rc0 v2.0.0-beta0 v2.0.0-alpha0 v1.8.5 v1.8.4 v1.8.3 v1.8.2 v1.8.1 v1.8.0 v1.7.2 v1.7.1 v1.7.0 v1.6.3 v1.6.2 v1.6.1 v1.6.0 v1.6.0-rc0 v1.5.2 v1.5.1 v1.5.0 v1.4.1 v1.4.0 v1.3.2 v1.3.1 v1.3.0 v1.2.1 v1.2.0 v1.1.0 v1.0.2 v1.0.1 v1.0.0 v1.0.0-rc0 v0.15.0 v0.15.0-rc0 v0.14.0 lite-v0.1
无相关合并请求
......@@ -248,7 +248,8 @@ void ListenAndServOp::RunImpl(const framework::Scope &scope,
request_prefetch_handler_.get());
auto *optimize_block = Attr<framework::BlockDesc *>(kOptimizeBlock);
auto *prefetch_block = Attr<framework::BlockDesc *>(kPrefetchBlock);
auto grad_to_block_id_str = Attr<std::vector<std::string>>(kPrefetchBlock);
framework::BlockDesc *prefetch_block = nullptr;
auto *program = optimize_block->Program();
framework::Executor executor(dev_place);
......@@ -302,8 +303,8 @@ class ListenAndServOpMaker : public framework::OpProtoAndCheckerMaker {
AddAttr<bool>("sync_mode", "if works at sync_mode or not").SetDefault(true);
AddAttr<framework::BlockDesc *>(kOptimizeBlock,
"BlockID to run on server side.");
AddAttr<framework::BlockDesc *>(kPrefetchBlock,
"prefetch block to run on server side.");
AddAttr<std::vector<std::string>>(kPrefetchBlock,
"prefetch block to run on server side.");
AddAttr<int>("Fanin", "How many clients send to this server.")
.SetDefault(1);
}
......
......@@ -30,7 +30,7 @@ namespace paddle {
namespace operators {
constexpr char kOptimizeBlock[] = "OptimizeBlock";
constexpr char kPrefetchBlock[] = "PrefetchBlock";
constexpr char kPrefetchBlock[] = "prefetch_var_name_to_block_id";
void RunServer(std::shared_ptr<detail::RPCServer> service);
......
......@@ -515,21 +515,20 @@ class DistributeTranspiler:
grad_to_block_id, None)
# process distributed lookup_table
prefetch_block = None
prefetch_var_name_to_block_id = []
if self.has_distributed_lookup_table:
pserver_index = self.pserver_endpoints.index(endpoint)
table_opt_block = self._create_table_optimize_block(
pserver_index, pserver_program, pre_block_idx, grad_to_block_id)
prefetch_block = self._create_prefetch_block(
prefetch_var_name_to_block_id = self._create_prefetch_block(
pserver_index, pserver_program, table_opt_block)
# NOTE: if has_distributed_lookup_table is False, then prefetch_block will
# not be executed, so it's safe to use optimize_block to hold the place
if self.has_distributed_lookup_table:
assert prefetch_block is not None
assert len(prefetch_var_name_to_block_id) > 0
else:
assert prefetch_block is None
prefetch_block = pserver_program.global_block()
assert len(prefetch_var_name_to_block_id) == 0
# step5 append the listen_and_serv op
pserver_program.global_block().append_op(
......@@ -540,7 +539,7 @@ class DistributeTranspiler:
"OptimizeBlock": pserver_program.block(1),
"endpoint": endpoint,
"Fanin": self.trainer_num,
"PrefetchBlock": prefetch_block,
"prefetch_var_name_to_block_id": prefetch_var_name_to_block_id,
"sync_mode": self.sync_mode,
"grad_to_block_id": grad_to_block_id
})
......@@ -608,8 +607,15 @@ class DistributeTranspiler:
def _replace_lookup_table_op_with_prefetch(self, program,
pserver_endpoints):
# 1. replace lookup_table_op with split_ids_op -> prefetch_op -> sum_op
self.prefetch_input_vars = None
self.prefetch_output_vars = None
# self.all_prefetch_input_vars =
# [[var0_prefetch_in_pserver0, var0_prefetch_in_pserver1]
# [var1_prefetch_in_pserver0, var1_prefetch_in_pserver1]]
self.all_prefetch_input_vars = []
# self.all_prefetch_input_vars =
# [[var0_prefetch_in_pserver0, var0_prefetch_in_pserver1]
# [var1_prefetch_in_pserver0, var1_prefetch_in_pserver1]]
self.all_prefetch_output_vars = []
continue_search_lookup_table_op = True
while continue_search_lookup_table_op:
......@@ -623,18 +629,19 @@ class DistributeTranspiler:
ids_name = op.input("Ids")
out_name = op.output("Out")
if self.prefetch_input_vars is None:
ids_var = program.global_block().vars[ids_name[0]]
self.prefetch_input_vars = self.create_splited_vars(
source_var=ids_var,
block=program.global_block(),
tag="_prefetch_in_")
if self.prefetch_output_vars is None:
out_var = program.global_block().vars[out_name[0]]
self.prefetch_output_vars = self.create_splited_vars(
source_var=out_var,
block=program.global_block(),
tag="_prefetch_out_")
ids_var = program.global_block().vars[ids_name[0]]
prefetch_input_vars = self.create_splited_vars(
source_var=ids_var,
block=program.global_block(),
tag="_prefetch_in_")
self.all_prefetch_input_vars.append(prefetch_input_vars)
out_var = program.global_block().vars[out_name[0]]
prefetch_output_vars = self.create_splited_vars(
source_var=out_var,
block=program.global_block(),
tag="_prefetch_out_")
self.all_prefetch_output_vars.append(prefetch_output_vars)
# insert split_ids_op
program.global_block().insert_op(
......@@ -646,14 +653,14 @@ class DistributeTranspiler:
for varname in ids_name
]
},
outputs={"Out": self.prefetch_input_vars})
outputs={"Out": prefetch_input_vars})
# insert prefetch_op
program.global_block().insert_op(
index=op_index + 1,
type="prefetch",
inputs={'X': self.prefetch_input_vars},
outputs={"Out": self.prefetch_output_vars},
inputs={'X': prefetch_input_vars},
outputs={"Out": prefetch_output_vars},
attrs={
"epmap": pserver_endpoints,
RPC_OP_ROLE_ATTR_NAME: RPC_OP_ROLE_ATTR_VALUE
......@@ -663,7 +670,7 @@ class DistributeTranspiler:
program.global_block().insert_op(
index=op_index + 2,
type="concat",
inputs={'X': self.prefetch_output_vars},
inputs={'X': prefetch_output_vars},
outputs={
"Out": [
program.global_block().vars[varname]
......@@ -709,30 +716,34 @@ class DistributeTranspiler:
optimize_block):
# STEP: create prefetch block
table_var = pserver_program.global_block().vars[self.table_name]
prefetch_block = pserver_program.create_block(optimize_block.idx)
trainer_ids = self.prefetch_input_vars[pserver_index]
pserver_ids = pserver_program.global_block().create_var(
name=trainer_ids.name,
type=trainer_ids.type,
shape=trainer_ids.shape,
dtype=trainer_ids.dtype)
trainer_out = self.prefetch_output_vars[pserver_index]
pserver_out = pserver_program.global_block().create_var(
name=trainer_out.name,
type=trainer_out.type,
shape=trainer_out.shape,
dtype=trainer_out.dtype)
prefetch_block.append_op(
type="lookup_sparse_table",
inputs={'Ids': pserver_ids,
"W": table_var},
outputs={"Out": pserver_out},
attrs={
"is_sparse": True, # has no effect on lookup_table op
"is_distributed": True,
"padding_idx": -1
})
return prefetch_block
prefetch_var_name_to_block_id = []
for index in range(len(self.all_prefetch_input_vars)):
prefetch_block = pserver_program.create_block(optimize_block.idx)
trainer_ids = self.all_prefetch_input_vars[index][pserver_index]
pserver_ids = pserver_program.global_block().create_var(
name=trainer_ids.name,
type=trainer_ids.type,
shape=trainer_ids.shape,
dtype=trainer_ids.dtype)
trainer_out = self.all_prefetch_output_vars[index][pserver_index]
pserver_out = pserver_program.global_block().create_var(
name=trainer_out.name,
type=trainer_out.type,
shape=trainer_out.shape,
dtype=trainer_out.dtype)
prefetch_block.append_op(
type="lookup_sparse_table",
inputs={'Ids': pserver_ids,
"W": table_var},
outputs={"Out": pserver_out},
attrs={
"is_sparse": True, # has no effect on lookup_table op
"is_distributed": True,
"padding_idx": -1
})
prefetch_var_name_to_block_id.append(trainer_ids.name + ":" + str(
prefetch_block.idx))
return prefetch_var_name_to_block_id
def _create_table_optimize_block(self, pserver_index, pserver_program,
pre_block_idx, grad_to_block_id):
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册