Serialization trade-offs in distributed systems: A comparative analysis of JSON, Protocol Buffers, and FlatBuffers under varying width and nesting depth

Johannes Haberlah and Knut Linke

IU Discussion Papers - IT & Engineering from IU International University of Applied Sciences

Abstract: This discussion paper compares three common serialization paradigms - JSON, Protocol Buffers, and FlatBuffers - with respect to latency and space efficiency. Although serialization is central to modern distributed systems, controlled, paradigm-level comparisons remain limited. To address this, the study benchmarks all three formats in Rust while systematically varying record width and nesting depth. Quantitative measurements are complemented by CPU profiling to identify dominant cost drivers. Results show clear patterns: Protocol Buffers achieves the lowest serialization latency and smallest message sizes, FlatBuffers offers the fastest data access due to its zero-copy design, and JSON incurs the highest access costs and largest payloads. Each format therefore excels under different conditions: Protobuf in write-and bandwidth-sensitive scenarios, FlatBuffers in read-dominated contexts, and JSON where readability and flexibility matter.

Keywords: Serialization; Protocol Buffers; FlatBuffers; Zero-copy; Benchmarking (search for similar items in EconPapers)
JEL-codes: C80 C88 L86 (search for similar items in EconPapers)
Date: 2026
New Economics Papers: this item is included in nep-inv
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.econstor.eu/bitstream/10419/343093/1/1981783016.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:zbw:iubhit:343093

DOI: 10.56250/4143

Access Statistics for this paper

More papers in IU Discussion Papers - IT & Engineering from IU International University of Applied Sciences
Bibliographic data for series maintained by ZBW - Leibniz Information Centre for Economics ().