ElasticRec: A Microservice-based Model Serving Architecture Enabling Elastic Resource Scaling for Recommendation Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Choi, Yujeong, Kim, Jiin, Rhu, Minsoo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914831251013632
author Choi, Yujeong
Kim, Jiin
Rhu, Minsoo
author_facet Choi, Yujeong
Kim, Jiin
Rhu, Minsoo
contents With the increasing popularity of recommendation systems (RecSys), the demand for compute resources in datacenters has surged. However, the model-wise resource allocation employed in current RecSys model serving architectures falls short in effectively utilizing resources, leading to sub-optimal total cost of ownership. We propose ElasticRec, a model serving architecture for RecSys providing resource elasticity and high memory efficiency. ElasticRec is based on a microservice-based software architecture for fine-grained resource allocation, tailored to the heterogeneous resource demands of RecSys. Additionally, ElasticRec achieves high memory efficiency via our utility-based resource allocation. Overall, ElasticRec achieves an average 3.3x reduction in memory allocation size and 8.1x increase in memory utility, resulting in an average 1.6x reduction in deployment cost compared to state-of-the-art RecSys inference serving system.
format Preprint
id arxiv_https___arxiv_org_abs_2406_06955
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ElasticRec: A Microservice-based Model Serving Architecture Enabling Elastic Resource Scaling for Recommendation Models
Choi, Yujeong
Kim, Jiin
Rhu, Minsoo
Distributed, Parallel, and Cluster Computing
Information Retrieval
Machine Learning
With the increasing popularity of recommendation systems (RecSys), the demand for compute resources in datacenters has surged. However, the model-wise resource allocation employed in current RecSys model serving architectures falls short in effectively utilizing resources, leading to sub-optimal total cost of ownership. We propose ElasticRec, a model serving architecture for RecSys providing resource elasticity and high memory efficiency. ElasticRec is based on a microservice-based software architecture for fine-grained resource allocation, tailored to the heterogeneous resource demands of RecSys. Additionally, ElasticRec achieves high memory efficiency via our utility-based resource allocation. Overall, ElasticRec achieves an average 3.3x reduction in memory allocation size and 8.1x increase in memory utility, resulting in an average 1.6x reduction in deployment cost compared to state-of-the-art RecSys inference serving system.
title ElasticRec: A Microservice-based Model Serving Architecture Enabling Elastic Resource Scaling for Recommendation Models
topic Distributed, Parallel, and Cluster Computing
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2406.06955