A performance comparison of web programming interfaces: GraphQL, gRPC and Thrift

Key findings from this study

This research indicates that:

• gRPC achieves the fastest execution time for large-scale read operations
• GraphQL and Thrift outperform gRPC when handling operations with small data volumes
• Performance trade-offs among these technologies vary significantly based on operation type and data transfer requirements

Overview

This study evaluates the performance characteristics of three web interface technologies: GraphQL, gRPC, and Thrift. The comparison measures execution time and data volume across database operations to identify performance trade-offs among these commonly deployed technologies.

Methods and approach

The researchers developed three separate applications, each implementing one of the target technologies. JMeter conducted performance testing across all implementations. Measurements captured execution time and data volume returned during relational database operations.

Results

gRPC demonstrated the fastest execution times for large-scale read operations. GraphQL and Thrift performed more efficiently when transporting small data volumes. The performance differential reflects the architectural differences in how each technology handles query specificity, serialization overhead, and data transmission.

Implications

Technology selection for web interface implementation should depend on anticipated workload characteristics and data transfer patterns. Systems expecting large-scale read operations benefit from gRPC's performance advantages. Applications optimized for minimal data payloads may achieve better results with GraphQL or Thrift despite other architectural considerations.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.