zakura-trace-plots

You can diagnose Zakura sync slowdowns by generating time-series plots that correlate block sync states with throughput. This skill automates the creation of visual metrics from raw trace data to help identify the root cause of node performance degradation.

You can identify Head-of-Line stalls in Zakura tracing data by running this skill to generate automated diagnostic summary reports. It analyzes feedrun CSV files to detect HoL stalls and commit phase memory pressure signatures directly from the trace directory.

Yes, you need Python 3 installed to generate performance plots from JSONL and CSV data. This skill relies on Python 3 standard library modules to process block_sync.jsonl and feedrun CSV files for performance visualization.

You diagnose throughput bottlenecks from block_sync.jsonl files by parsing the trace data to generate metrics-aware SVG plots. This skill analyzes applying and reorder metrics within the JSONL data to pinpoint the exact sync slowdowns.

Yes, you can analyze commit phase memory pressure from Zakura trace directories without external dependencies. This skill requires only Python 3 standard library modules to process JSONL and CSV data, requiring no external packages to generate diagnostic summaries.

Zakura node sync slowdowns during block processing often stem from throughput bottlenecks, HoL stalls, or commit phase memory pressure. This skill helps identify the root cause by generating time-series plots that correlate block sync states with throughput metrics.