Delivering Document Conversion as a Cloud Service with High Throughput and Responsiveness

TL;DR

This paper presents a scalable cloud-based document conversion service capable of processing over one million PDF pages per hour by optimizing workload distribution and resource management.

Contribution

It introduces a novel scalable architecture for document conversion in the cloud, addressing challenges of high throughput and responsiveness for complex, variable document formats.

Findings

Achieved over one million pages per hour throughput

Compared two workload distribution strategies and configurations

Demonstrated high resource efficiency and scalability

Abstract

Document understanding is a key business process in the data-driven economy since documents are central to knowledge discovery and business insights. Converting documents into a machine-processable format is a particular challenge here due to their huge variability in formats and complex structure. Accordingly, many algorithms and machine-learning methods emerged to solve particular tasks such as Optical Character Recognition (OCR), layout analysis, table-structure recovery, figure understanding, etc. We observe the adoption of such methods in document understanding solutions offered by all major cloud providers. Yet, publications outlining how such services are designed and optimized to scale in the cloud are scarce. In this paper, we focus on the case of document conversion to illustrate the particular challenges of scaling a complex data processing pipeline with a strong reliance on machine-learning methods on cloud infrastructure. Our key objective is to achieve high scalability and responsiveness for different workload profiles in a well-defined resource budget. We outline the requirements, design, and implementation choices of our document conversion service and reflect on the challenges we faced. Evidence for the scaling behavior and resource efficiency is provided for two alternative workload distribution strategies and deployment configurations. Our best-performing method achieves sustained throughput of over one million PDF pages per hour on 3072 CPU cores across 192 nodes.