QUIC Spin Bit Values for RTT Analysis from IMC '23 Study

A dataset from RWTH Aachen University researchers contains extracted raw spin bit values used to analyze the adoption and use of QUIC's spin bit feature. The data, supporting a 2023 Internet Measurement Conference paper, includes timestamps, spin bit values, packet numbers, and RTT estimates from client-side measurements. It is intended for analysis with provided Go scripts to detect spin modes and calculate RTTs.

Use Cases

• Detecting spin modes (on/off/undefined) based on the sequence of sent and received spin bit values.
• Calculating round-trip times based on spin bit measurements and comparing them to RTTs from the QUIC stack.
• Analyzing packet reordering effects on RTT calculations by using the provided packet numbers.
• Studying the adoption rate of the QUIC spin bit feature in the wild from the collected traffic traces.

Strengths

• Includes raw data (SentSpinBits, RecvSpinBits, QlogRTTs) used in a peer-reviewed ACM IMC 2023 publication.
• Provides companion Go analysis scripts for reproducibility, allowing detection of spin modes and RTT calculation.
• Data format is specified as line-wise dictionaries, suggesting a structured, machine-readable format.

Limitations

• Row count, total size, and specific file formats are unknown, limiting suitability assessment for large-scale processing.
• Column-level documentation is absent for the final analysis output; field semantics must be inferred from the description.
• Last update date is unknown; freshness of the data collection is unverified.

Provenance

Source

Ike Kunze, Constantin Sander, Klaus Wehrle (RWTH Aachen University)

Collection Method

Extracted from client-side measurements of QUIC traffic, as described in the associated IMC '23 paper.