CRISP: Critical Path Analysis of Microservice Traces

This repo contains code to compute and present critical path summary from Jaeger microservice traces. To use first collect the microservice traces of a specific endpoint in a directory (say traces). Let the traces be for OP operation and SVC service (these are Jaeger termonologies). python3 process.py --operationName OP --serviceName SVC -t <path to trace> -o . --parallelism 8 will produce the critical path summary using 8 concurrent processes. The summary will be output in the current directory as an HTML file with a heatmap, flamegraph, and summary text in criticalPaths.html. It will also produce three flamegraphs flame-graph-*.svg for three different percentile values.

The script accepts the following options:

python3 process.py --help
usage: process.py [-h] -a OPERATIONNAME -s SERVICENAME [-t TRACEDIR] [--file FILE] -o OUTPUTDIR
                  [--parallelism PARALLELISM] [--topN TOPN] [--numTrace NUMTRACE] [--numOperation NUMOPERATION]

optional arguments:
  -h, --help            show this help message and exit
  -a OPERATIONNAME, --operationName OPERATIONNAME
                        operation name
  -s SERVICENAME, --serviceName SERVICENAME
                        name of the service
  -t TRACEDIR, --traceDir TRACEDIR
                        path of the trace directory (mutually exclusive with --file)
  --file FILE           input path of the trace file (mutually exclusivbe with --traceDir)
  -o OUTPUTDIR, --outputDir OUTPUTDIR
                        directory where output will be produced
  --parallelism PARALLELISM
                        number of concurrent python processes.
  --topN TOPN           number of services to show in the summary
  --numTrace NUMTRACE   number of traces to show in the heatmap
  --numOperation NUMOPERATION
                        number of operations to show in the heatmap

CRISP: Critical Path Analysis of Microservice Traces

Related tags

Overview

CRISP: Critical Path Analysis of Microservice Traces

Owner

Uber Research

BinTuner is a cost-efficient auto-tuning framework, which can deliver a near-optimal binary code that reveals much more differences than -Ox settings.

Incubator for useful bioinformatics code, primarily in Python and R

Full automated data pipeline using docker images

A real data analysis and modeling project - restaurant inspections

A simple and efficient tool to parallelize Pandas operations on all available CPUs

MEAD: A Large-scale Audio-visual Dataset for Emotional Talking-face Generation [ECCV2020]

Business Intelligence (BI) in Python, OLAP

Pyspark Spotify ETL

MIR Cheatsheet - Survival Guidebook for MIR Researchers in the Lab

This is an analysis and prediction project for house prices in King County, USA based on certain features of the house

Show you how to integrate Zeppelin with Airflow

Gaussian processes in TensorFlow

Titanic data analysis for python

HyperSpy is an open source Python library for the interactive analysis of multidimensional datasets

Python data processing, analysis, visualization, and data operations

A collection of learning outcomes data analysis using Python and SQL, from DQLab.

A set of tools to analyse the output from TraDIS analyses

Investigating EV charging data

Datashader is a data rasterization pipeline for automating the process of creating meaningful representations of large amounts of data.

The micro-framework to create dataframes from functions.