【发布时间】:2019-09-13 07:51:41
【问题描述】:
我目前还不熟悉在 Python 中将 Apache Beam 与 Dataflow runner 一起使用。我有兴趣创建一个发布到 Google Cloud PubSub 的批处理管道,我已经修改了 Beam Python API 并找到了一个解决方案。然而,在我的探索过程中,我遇到了一些有趣的问题,让我很好奇。
1。成功的管道
目前,我从 GCS 批量发布数据的成功光束管道如下所示:
class PublishFn(beam.DoFn):
def __init__(self, topic_path):
self.topic_path = topic_path
super(self.__class__, self).__init__()
def process(self, element, **kwargs):
from google.cloud import pubsub_v1
publisher = pubsub_v1.PublisherClient()
future = publisher.publish(self.topic_path, data=element.encode("utf-8"))
return future.result()
def run_gcs_to_pubsub(argv):
options = PipelineOptions(flags=argv)
from datapipes.common.dataflow_utils import CsvFileSource
from datapipes.protos import proto_schemas_pb2
from google.protobuf.json_format import MessageToJson
with beam.Pipeline(options=options) as p:
normalized_data = (
p |
"Read CSV from GCS" >> beam.io.Read(CsvFileSource(
"gs://bucket/path/to/file.csv")) |
"Normalize to Proto Schema" >> beam.Map(
lambda data: MessageToJson(
proto_schemas_pb2(data, proto_schemas_pb2.MySchema()),
indent=0,
preserving_proto_field_name=True)
)
)
(normalized_data |
"Write to PubSub" >> beam.ParDo(
PublishFn(topic_path="projects/my-gcp-project/topics/mytopic"))
)
2。不成功的管道
在这里,我试图让发布者在DoFn 之间共享。我尝试了以下方法。
一个。在 DoFn 中初始化发布者
class PublishFn(beam.DoFn):
def __init__(self, topic_path):
from google.cloud import pubsub_v1
batch_settings = pubsub_v1.types.BatchSettings(
max_bytes=1024, # One kilobyte
max_latency=1, # One second
)
self.publisher = pubsub_v1.PublisherClient(batch_settings)
self.topic_path = topic_path
super(self.__class__, self).__init__()
def process(self, element, **kwargs):
future = self.publisher.publish(self.topic_path, data=element.encode("utf-8"))
return future.result()
def run_gcs_to_pubsub(argv):
... ## same as 1
b.在 DoFn 之外初始化 Publisher,并将其传递给 DoFn
class PublishFn(beam.DoFn):
def __init__(self, publisher, topic_path):
self.publisher = publisher
self.topic_path = topic_path
super(self.__class__, self).__init__()
def process(self, element, **kwargs):
future = self.publisher.publish(self.topic_path, data=element.encode("utf-8"))
return future.result()
def run_gcs_to_pubsub(argv):
.... ## same as 1
batch_settings = pubsub_v1.types.BatchSettings(
max_bytes=1024, # One kilobyte
max_latency=1, # One second
)
publisher = pubsub_v1.PublisherClient(batch_settings)
with beam.Pipeline(options=options) as p:
... # same as 1
(normalized_data |
"Write to PubSub" >> beam.ParDo(
PublishFn(publisher=publisher, topic_path="projects/my-gcp-project/topics/mytopic"))
)
使发布者在DoFn 方法之间共享的两次尝试均失败,并显示以下错误消息:
File "stringsource", line 2, in grpc._cython.cygrpc.Channel.__reduce_cython__
和
File "stringsource", line 2, in grpc._cython.cygrpc.Channel.__reduce_cython__
TypeError: no default __reduce__ due to non-trivial __cinit__
我的问题是:
共享发布者的实施会提高光束管道的性能吗?如果是,那么我想探索这个解决方案。
为什么我的失败管道会出现错误?是因为在
process函数之外初始化自定义类对象并将其传递给 DoFn 吗?如果是由于这个原因,我该如何实现一个管道,以便能够在 DoFn 中重用自定义对象?
谢谢您,我们将不胜感激。
编辑:解决方案
好的,所以 Ankur 已经解释了为什么会出现我的问题,并讨论了如何在 DoFn 上进行序列化。基于这些知识,我现在了解到有两种解决方案可以在 DoFn 中使自定义对象共享/可重用:
使自定义对象可序列化:这允许对象在 DoFn 对象创建期间被初始化/可用(在
__init__下)。该对象必须是可序列化的,因为它将在创建 DoFn 对象的管道提交期间被序列化(调用__init__)。如何实现这一点在我的回答中得到了回答。此外,我发现此要求实际上与 [1][2] 下的 Beam 文档相关联。在
__init__之外的 DoFn 函数中初始化不可序列化对象以避免序列化,因为在管道提交期间不会调用 init 之外的函数。 Ankur 的回答中解释了如何做到这一点。
参考资料:
[1]https://beam.apache.org/documentation/programming-guide/#core-beam-transforms
【问题讨论】:
标签: python google-cloud-dataflow apache-beam