Migrating a Recommendation System to Cloud Using ML Workflow

When using an ML deployment service such as SageMaker, the service may handle the deployment of serialized models.

“When using an ML deployment service (e.g., SageMaker 7 ), the service may handle the deployment of serialized models [9].”

Another reported method of integration is to serialize the model and load it at runtime, possibly with an ML framework optimized for production.

“Another reported method of integration is to serialize the model (e.g., into ONNXfoot_2 , HDF5, Joblib 4 ) and load it at runtime, possibly with an ML framework that is optimized for production [9,21,39,40].”

Serving-related challenges are some of the ones most often reported in the literature, usually from a performance perspective.

“Serving-related challenges are some of the ones most often reported in the literature [5-7, 9, 11, 12, 16, 20, 28, 34, 44, 53, 54], usually from a performance perspective.”

Increasing hardware instances and network bandwidth both improve inference latencies.

“The difficulty in achieving cost-effective and performant inference was also corroborated by [9], in which the authors observed that increasing hardware instances and network bandwidth both improve inference latencies.”

A static storage service such as AWS S3 may be used for model storage when models are downloaded by the target deployment environment.

“When models are downloaded by the target deployment environment, a static storage service such as AWS S3 27 may be used for model storage [9,29].”

Polyaxon, MLflow, TFX, ZenML, Flyte, Seldon Core, BentoML, and Amazon SageMaker are among the more fully integrated platforms for operationalizing ML.

“Polyaxon 9 , MLflow 10 , TFX 11 , ZenML 12 , Flyte 13 , Seldon Core 14 and BentoML 15 are some of the more fully-integrated platforms that exist [10,45], along with the Amazon-exclusive Sage-Maker 16 [9,45].”