• Stars
    star
    1,541
  • Rank 29,121 (Top 0.6 %)
  • Language
    Rust
  • License
    Apache License 2.0
  • Created about 5 years ago
  • Updated 12 months ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Systems performance telemetry

Rezolus

Rezolus is a tool for collecting detailed systems performance telemetry and exposing burst patterns through high-resolution telemetry. Rezolus provides instrumentation of basic systems metrics, performance counters, and support for eBPF (extended Berkeley Packet Filter) telemetry. Measurement is the first step toward improved performance.

Per-metric documentation can be found in the METRICS documentation.

License: Apache-2.0 Build Status: CI

Overview

Rezolus collects telemetry from several different sources. Currently, Rezolus collects telemetry from traditional sources (procfs, sysfs), the perf_events subsystem, and from BPF. Each sampler implements a consistent set of functions so that new ones can be easily added to further extend the capabilities of Rezolus.

Each telemetry source is oversampled so that we can build a histogram across a time interval. This histogram allows us to capture variations which will appear in the far upper and lower percentiles. This oversampling approach is one of the key differentiators of Rezolus when compared to other telemetry agents.

With its support for BPF as well as more common telemetry sources, Rezolus is a very sophisticated tool for capturing performance anomalies, profiling systems performance, and conducting performance diagnostics.

More detailed information about the underlying metrics library and sampler design can be found in the DESIGN documentation.

Features

  • traditional telemetry sources (procfs, sysfs, ...)
  • perf_events support for hardware performance counters
  • BPF support to instrument kernel and user space activities
  • oversampling and percentile metrics to capture bursts

Traditional Telemetry Sources

Rezolus collects metrics from traditional sources (procfs, sysfs) to provide basic telemetry for CPU, disk, and network. Rezolus exports CPU utilization, disk bandwidth, disk IOPs, network bandwidth, network packet rate, network errors, as well as TCP and UDP protocol counters.

These basic telemetry sources, when coupled with the approach of oversampling to capture their bursts, often provide a high-level view of systems performance and may readily indicate areas where resources are saturated or errors are occurring.

Perf Events

Perf Events allow us to report on both hardware and software events. Typical software events are things like page faults, context switches, and CPU migrations. Typical hardware events are things like CPU cycles, instructions retired, cache hits, cache misses, and a variety of other detailed metrics about how a workload is running on the underlying hardware.

These metrics are typically used for advanced performance debugging, as well as for tuning and optimization efforts.

BPF

There is an expansive amount of performance information that can be exposed through BPF, which allows us to have the Linux Kernel perform telemetry capture and aggregation at very fine-grained levels.

Rezolus comes with samplers that capture block IO size distribution, EXT4 and XFS operation latency distribution, and scheduler run queue latency distribution. You'll see that here we are mainly exposing distributions of sizes and latencies The kernel is recording the appropriate value for each operation into a histogram. Rezolus then accesses this histogram from user-space and transfers the values over to its own internal storage where it is then exposed to external aggregators.

By collecting telemetry in-kernel, we're able to gather data about events that happen at extremely high rates - e.g., task scheduling - with minimal performance overhead for collecting the telemetry. The BPF samplers can be used to both capture runtime performance anomalies as well as characterize workloads.

Sampling rate and resolution

In order to accurately reflect the intensity of a burst, the sampling rate must be at least twice the duration of the shortest burst to record accurately. This ensures that at least 1 sample completely overlaps the burst section of the event. With a traditional minutely time series, this means that a spike must least 120 seconds or more to be accurately recorded in terms of intensity. Rezolus allows for sampling rate to be configured, allowing us to make a trade-off between resolution and resource consumption. At 10Hz sampling, 200ms or more of consecutive burst is enough to be accurately reflected in the pMax. Contrast that with minutely metrics requiring 120_000ms, or secondly requiring 2000ms of consecutive burst to be accurately recorded.

Getting Started

Building

Rezolus is built with the standard Rust toolchain which can be installed and managed via rustup or by following the directions on the Rust website.

The rest of the guide assumes you've chosen to install the toolchain via rustup.

NOTE: Rezolus is intended to be built and deployed on Linux systems but has some very limited support for MacOS to test the overall framework. It is focused on providing systems telemetry for Linux systems. To get the best experience and develop new samplers, you should build and run on Linux.

Clone and build Rezolus from source

git clone https://github.com/twitter/rezolus
cd rezolus

# create an unoptimized development build
cargo build

# run the unoptimized binary and display help
cargo run -- --help

# create an optimized release build
cargo build --release

# run the optimized binary and display help
cargo run --release -- --help

# run the optimized binary with the example config (needs sudo for perf_events)
cargo build --release && \
sudo target/release/rezolus --config configs/example.toml

# metrics can be viewed in human-readable form with curl
curl --silent http://localhost:4242/vars

Building with BPF Support

By default, BPF support is not compiled in. If you wish to produce a build with BPF support enabled, follow the steps below:

Prerequisites

BPF support requires that we link against the BPF Compiler Collection. You may either use the version provided by your distribution, or can build BCC and install from source. It is critical to know which version of BCC you have installed. Rezolus supports multiple versions by utilizing different feature flags at build time.

Our current policy is to support BCC versions back to the version in the distribution repository for Debian Stable or CentOS 7 (whichever is older) to the most recent version of BCC.

This policy provides coverage for current stable and testing versions of Debian, Ubuntu, CentOS, Fedora, Arch, and Gentoo. Users of other distributions may need to build a supported version of BCC from source. See BCC Installation Guide for details.

Building

As mentioned above, we provide different feature flags to map to various supported BCC versions. The bpf feature will map to the newest version supported by the rust-bcc project. For most users, this will be the right flag to use. However, if you must link against an older BCC version, you will need to use a more specific form of the feature flag. These version-specific flags take the form of bpf_v0_10_0 with the BCC version being reflected in the name of the feature. You can find a complete list of the feature flags in the cargo manifest for this project.

# create an optimized release build with BPF support
cargo build --release --features bpf
sudo target/release/rezolus --config configs/example.toml

# metrics can be viewed in human-readable form with curl
curl --silent http://localhost:4242/vars

HTTP Exposition

Rezolus exposes metrics over HTTP, with different paths corresponding to different exposition formats.

  • human-readable: /vars
  • JSON: /vars.json, /metrics.json, /admin/metrics.json
  • Prometheus: /metrics

NOTE: currently, JSON exposition is provided by default for any other path. This behavior may change in the future and should not be relied on.

Additionally, you can get the running version on the root-level path /

Support

Create a new issue on GitHub.

Contributing

We feel that a welcoming community is important and we ask that you follow Twitter's Open Source Code of Conduct in all interactions with the community.

Authors

A full list of contributors can be found on GitHub.

Follow @TwitterOSS on Twitter for updates.

License

Copyright 2019 Twitter, Inc.

Licensed under the Apache License, Version 2.0: https://www.apache.org/licenses/LICENSE-2.0

Security Issues?

Please report sensitive security issues via Twitter's bug-bounty program (https://hackerone.com/twitter) rather than GitHub.

More Repositories

1

the-algorithm

Source code for Twitter's Recommendation Algorithm
Scala
60,968
star
2

twemoji

Emoji for everyone. https://twemoji.twitter.com/
HTML
16,575
star
3

typeahead.js

typeahead.js is a fast and fully-featured autocomplete library
JavaScript
16,526
star
4

twemproxy

A fast, light-weight proxy for memcached and redis
C
12,000
star
5

the-algorithm-ml

Source code for Twitter's Recommendation Algorithm
Python
9,844
star
6

finagle

A fault tolerant, protocol-agnostic RPC system
Scala
8,742
star
7

hogan.js

A compiler for the Mustache templating language
JavaScript
5,143
star
8

labella.js

Placing labels on a timeline without overlap.
JavaScript
3,869
star
9

scala_school

Lessons in the Fundamentals of Scala
HTML
3,700
star
10

AnomalyDetection

Anomaly Detection with R
R
3,529
star
11

scalding

A Scala API for Cascading
Scala
3,467
star
12

twitter-text

Twitter Text Libraries. This code is used at Twitter to tokenize and parse text to meet the expectations for what can be used on the platform.
HTML
3,051
star
13

TwitterTextEditor

A standalone, flexible API that provides a full-featured rich text editor for iOS applications.
Swift
2,950
star
14

opensource-website

Twitter's open source website, identifying projects we've released, organizations we support, and the work we do to support open source.
SCSS
2,918
star
15

util

Wonderful reusable code from Twitter
Scala
2,679
star
16

algebird

Abstract Algebra for Scala
Scala
2,284
star
17

finatra

Fast, testable, Scala services built on TwitterServer and Finagle
Scala
2,271
star
18

effectivescala

Twitter's Effective Scala Guide
HTML
2,241
star
19

summingbird

Streaming MapReduce with Scalding and Storm
Scala
2,136
star
20

pelikan

Pelikan is Twitter's unified cache backend
C
1,921
star
21

ios-twitter-image-pipeline

Twitter Image Pipeline is a robust and performant image loading and caching framework for iOS clients
C
1,851
star
22

twurl

OAuth-enabled curl for the Twitter API
Ruby
1,790
star
23

twitter-server

Twitter-Server defines a template from which services at Twitter are built
Scala
1,542
star
24

activerecord-reputation-system

An Active Record Reputation System for Rails
Ruby
1,334
star
25

communitynotes

Documentation and source code powering Twitter's Community Notes
Python
1,319
star
26

compose-rules

Static checks to aid with a healthy adoption of Compose
Kotlin
1,311
star
27

fatcache

Memcache on SSD
C
1,301
star
28

rsc

Experimental Scala compiler focused on compilation speed
Scala
1,245
star
29

elephant-bird

Twitter's collection of LZO and Protocol Buffer-related Hadoop, Pig, Hive, and HBase code.
Java
1,137
star
30

cassovary

Cassovary is a simple big graph processing library for the JVM
Scala
1,039
star
31

Serial

Light-weight, fast framework for object serialization in Java, with Android support.
Java
988
star
32

hbc

A Java HTTP client for consuming Twitter's realtime Streaming API
Java
963
star
33

twemcache

Twemcache is the Twitter Memcached
C
925
star
34

innovators-patent-agreement

Innovators Patent Agreement (IPA)
919
star
35

vireo

Vireo is a lightweight and versatile video processing library written in C++11
C++
919
star
36

twitter-korean-text

Korean tokenizer
Scala
834
star
37

scrooge

A Thrift parser/generator
Scala
785
star
38

BreakoutDetection

Breakout Detection via Robust E-Statistics
C++
753
star
39

GraphJet

GraphJet is a real-time graph processing library.
Java
696
star
40

twitter-cldr-rb

Ruby implementation of the ICU (International Components for Unicode) that uses the Common Locale Data Repository to format dates, plurals, and more.
Ruby
667
star
41

bijection

Reversible conversions between types
Scala
657
star
42

chill

Scala extensions for the Kryo serialization library
Scala
607
star
43

ios-twitter-network-layer

Twitter Network Layer is a scalable and feature rich network layer built on top of NSURLSession for Apple platforms
Objective-C
573
star
44

hadoop-lzo

Refactored version of code.google.com/hadoop-gpl-compression for hadoop 0.20
Shell
544
star
45

storehaus

Storehaus is a library that makes it easy to work with asynchronous key value stores
Scala
464
star
46

rpc-perf

A tool for benchmarking RPC services
Rust
458
star
47

d3kit

D3Kit is a set tools to speed D3 related project development
JavaScript
429
star
48

scoot

Scoot is a distributed task runner, supporting both a proprietary API and Bazel's Remote Execution.
Go
347
star
49

twitter-cldr-js

JavaScript implementation of the ICU (International Components for Unicode) that uses the Common Locale Data Repository to format dates, plurals, and more. Based on twitter-cldr-rb.
JavaScript
345
star
50

scala_school2

Scala School 2
Scala
340
star
51

rustcommon

Common Twitter Rust lib
Rust
339
star
52

wordpress

The official Twitter plugin for WordPress. Embed Twitter content and grow your audience on Twitter.
PHP
310
star
53

ios-twitter-logging-service

Twitter Logging Service is a robust and performant logging framework for iOS clients
Objective-C
299
star
54

nodes

A library to implement asynchronous dependency graphs for services in Java
Java
246
star
55

SentenTree

A novel text visualization technique
JavaScript
226
star
56

interactive

Twitter interactive visualization
HTML
213
star
57

joauth

A Java library for authenticating HTTP Requests using OAuth
Java
211
star
58

thrift_client

A Thrift client wrapper that encapsulates some common failover behavior
Ruby
196
star
59

hpack

Header Compression for HTTP/2
Java
192
star
60

zktraffic

ZooKeeper protocol analyzer and stats gathering daemon
Python
165
star
61

twemoji-parser

A simple library for identifying emoji entities within a string in order to render them as Twemoji.
Scala
162
star
62

cache-trace

A collection of Twitter's anonymized production cache traces.
Shell
162
star
63

sbf

Java
159
star
64

tormenta

Scala extensions for Storm
Scala
132
star
65

whiskey

HTTP library for Android (beta)
Java
131
star
66

hraven

hRaven collects run time data and statistics from MapReduce jobs in an easily queryable format
Java
127
star
67

netty-http2

HTTP/2 for Netty
Java
120
star
68

sqrl

A Safe, Stateful Rules Language for Event Streams
TypeScript
100
star
69

ccommon

Cache Commons
C
99
star
70

focus

Focus aligns Git worktree content based on outlines of a repository's Bazel build graph. Focused repos are sparse, shallow, and thin and unlock markedly better performance in large repos.
Rust
91
star
71

dict_minimize

Access scipy optimizers from your favorite deep learning framework.
Python
77
star
72

metrics

76
star
73

twitter.github.io

HTML
71
star
74

go-bindata

Go
68
star
75

diffusion-rl

Python
66
star
76

birdwatch

64
star
77

cloudhopper-commons

Cloudhopper Commons
Java
57
star
78

twitter-cldr-npm

TwitterCldr npm package
JavaScript
49
star
79

.github

Twitter GitHub Organization-wide files
48
star
80

bazel-multiversion

Bazel rules to resolve, fetch and manage 3rdparty JVM dependencies with support for multiple parallel versions of the same dependency. Powered by Coursier.
Scala
47
star
81

libwatchman

A C interface to watchman
C
44
star
82

sslconfig

Twitter's OpenSSL Configuration
42
star
83

ios-twitter-apache-thrift

A thrift encoding and decoding library for Swift
Swift
41
star
84

gatekeeper-service

GateKeeper is a service built to automate the manual steps involved in onboarding, offboarding, and lost asset scenarios.
Python
36
star
85

dodo

The Twitter OSS Project Builder
Shell
35
star
86

repo-scaffolding

Tools for creating repos based on open source standards and best practices
33
star
87

iago2

A load generator, built for engineers
Scala
24
star
88

caladrius

Performance modelling system for Distributed Stream Processing Systems (DSPS) such as Apache Heron and Apache Storm
Python
22
star
89

ossdecks

Repository for Twitter Open Source Decks
10
star
90

curation-style-guide

Document Repository for Twitter's Curation Style Guide
10
star
91

analytics-infra-governance

Description of the process for how to commit, review, and release code to the Scalding OSS family (Scalding, Summingbird, Algebird, Bijection, Storehaus, etc)
9
star
92

gpl-commitment

Twitter's GPL Cooperation Commitment
5
star
93

second-control-probability-distributions

4
star
94

google-tag-manager-event-tag

Smarty
3
star
95

google-tag-manager-base-tag

Smarty
2
star