https://github.com/prestodb/benchto

Framework for running macro benchmarks in a clustered environment
https://github.com/prestodb/benchto

Last synced: about 1 month ago
JSON representation

Framework for running macro benchmarks in a clustered environment

• Host: GitHub
• URL: https://github.com/prestodb/benchto
• Owner: prestodb
• License: apache-2.0
• Created: 2016-07-01T16:25:50.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2022-08-29T22:25:39.000Z (over 2 years ago)
• Last Synced: 2024-04-16T07:20:17.538Z (8 months ago)
• Language: JavaScript
• Homepage:
• Size: 3.62 MB
• Stars: 24
• Watchers: 100
• Forks: 19
• Open Issues: 24
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Benchto [![Build Status](https://travis-ci.org/prestodb/benchto.svg?branch=master)](https://travis-ci.org/prestodb/benchto)

The purpose of _Benchto_ project is to provide easy and manageable way to define, run and analyze _macro benchmarks_

in clustered environment. Understanding behaviour of distributed systems is hard and requires good visibility into

state of the cluster and internals of tested system. This project was developed for repeatable benchmarking of

Hadoop SQL engines, most importantly [Presto](https://prestodb.io/).

_Benchto_ consists of two main components: `benchto-service` and `benchto-driver`. To utilize all features of _Benchto_,

it is also recommended to configure _Graphite_ and _Grafana_. Image below depicts high level architecture:

![Benchto high level architecture](high-level-architecture.png?raw=true "Benchto high level architecture")

- _benchto-service_ - persistent data store for benchmark results. It exposes REST API and stores results in relational

DB (_Postgres_). Driver component calls API to store benchmark execution details which are later displayed by webapp

which is bundled with service. Single instance of _benchto-service_ can be shared by multiple benchmark drivers.

- _benchto-driver_ - standalone java application which loads definitions of benchmarks (_benchmark descriptors_) and

executes them against tested distributed system. If cluster monitoring is in place, driver collects various metrics

(cpu, memory, network usage) regarding cluster resource utilization. Moreover it adds graphite events which can be

later displayed as [Grafana annotations](http://docs.grafana.org/reference/annotations/). All data is stored in service

for later analysis.

- _monitoring_ - cluster monitoring is optional, but highly recommended to fully understand performance characteristics

of tested system. It is assumed that _Graphite/Carbon_ is used as metrics store and _Grafana_ for clusters dashboards.

There is no limitation on metric agents deployed on cluster hosts.

- _benchto-generator_ - map reduce job for generating benchmark data. You can configure number of rows, type of the

row (ex. _BIGINT_, _INT_, _DOUBLE_, _DECIMAL(38,8)_), output format (_ORC_, _TEXT_) and number of output files. More

details can be found in _benchto-generator_ module README file.