Performance Optimization in Ruby

Feb 10, 20170 likes262 views

It seems easier to scale the Heroku dynos or AWS servers with a slider, but is it really required? When developers see that their application is slow, their first reflex is to increase the dyno size. They hardly ever look at the application code. I've seen a lots of Ruby applications and most of them are over-scaled and wasting money. I intend to start by focusing on areas where Ruby takes a performance hit, such as Garbage collection and memory management. Then I will talk about the solutions to fix them in Ruby applications. Further I will showcase how why its better to use one function in Ruby over the other and the performance hit the application is taking because of them. These will be backed by the benchmarking suite I have created from my experience.

PERFORMANCE
OPTIMIZATION IN RUBYPRATHMESH RANAUT
? Prathmesh Ranaut, 2017

WHO AM I
> Computer Science Undergrad
> Helped ! of ruby apps improve
performance
? Prathmesh Ranaut, 2017

> Celluloid - Concurrent ruby framework
> Improved performance by 300%
? Prathmesh Ranaut, 2017

WHAT IS PERFORMANCE
OPTIMIZATION? !
? Prathmesh Ranaut, 2017

DOING
MOREWITH
LESS
? Prathmesh Ranaut, 2017

ALWAYS HAVE A BENCHMARKING SUITE
? Prathmesh Ranaut, 2017

require "benchmark/ips"
Benchmark.ips do |ips|
ips.report("addition") { 1 + 2 }
end
? Prathmesh Ranaut, 2017

require "benchmark/ips"
Benchmark.ips do |ips|
ips.report("addition") { 1 + 2 }
end
Warming up --------------------------------------
addition 325.015k i/100ms
Calculating -------------------------------------
addition 12.544M (± 8.5%) i/s - 62.403M in 5.012765s
? Prathmesh Ranaut, 2017

$class Worker include Celluloid def hashed(hash, key) hash[key] end end # Worker pool of different sizes pool_10 = Worker.pool(size: 10) pool_100 = Worker.pool(size: 100) hash = {} ENTRIES = 10_000 TESTS = 100_000 # Using a constant key, to derive consistent benchmarking results KEY = 500 # Populate hash array ENTRIES.times do |i| hash[i] = i end Benchmark.bmbm do |ips| puts "Finding the key : #{KEY}" ips.report("pool - 10") do TESTS.times do pool_10.async.hashed(hash, KEY) end end ips.report("pool - 100") do TESTS.times do pool_100.async.hashed(hash, KEY) end end end ? Prathmesh Ranaut, 2017$

RUBY 1.9
> YARV
? Prathmesh Ranaut, 2017

RUBY 1.9
> YARV
RUBY 2.0
> Copy-on-Write
? Prathmesh Ranaut, 2017

RUBY 2.1
> Generation GC
RUBY 2.4
> Improves the speed to access an instance variable
? Prathmesh Ranaut, 2017

TRY ANOTHER IMPLEMENTATION OF
RUBY
? Prathmesh Ranaut, 2017

RUBY 2.3 VS JRUBY
RUBY 2.3
Calculating -------------------------------------
addition 12.544M (± 8.5%) i/s - 62.403M
JRUBY
Calculating -------------------------------------
addition 22.929M (± 9.3%) i/s - 113.127M
? Prathmesh Ranaut, 2017

FIGURE OUT THE BOTTLENECK
I/O OR CPU OR MEMORY
? Prathmesh Ranaut, 2017

SWITCH TO ELIXIR
? Prathmesh Ranaut, 2017

TRY MULTITHREADING
? Prathmesh Ranaut, 2017

TOOLS
THAT MAKE MY LIFE EASIER
? Prathmesh Ranaut, 2017

METHOD PROFILER
HTTPS://GITHUB.COM/CHANGE/METHOD_PROFILER
? Prathmesh Ranaut, 2017

profiler = MethodProfiler.observe(BazingaClass)
# Do stuff with the BazingaClass
puts profiler.report
? Prathmesh Ranaut, 2017

KCACHEGRIND
HTTPS://KCACHEGRIND.GITHUB.IO/
? Prathmesh Ranaut, 2017

RUBYPROF
HTTPS://GITHUB.COM/RUBY-PROF/RUBY-PROF
? Prathmesh Ranaut, 2017

ObjectSpace.count_objects
? Prathmesh Ranaut, 2017

${ :TOTAL=>30165, :FREE=>2216, :T_OBJECT=>928, :T_CLASS=>574, :T_MODULE=>38, :T_FLOAT=>4, :T_STRING=>13182, :T_REGEXP=>77, :T_ARRAY=>2715, :T_HASH=>88, :T_STRUCT=>3, :T_BIGNUM=>2, :T_FILE=>30, :T_DATA=>363, :T_MATCH=>233, :T_COMPLEX=>1, :T_SYMBOL=>6, :T_IMEMO=>4530, :T_NODE=>5138, :T_ICLASS=>37 } ? Prathmesh Ranaut, 2017$

THANK YOU
&
QUESTIONSHELLO@PRATHMESHRANAUT.COM
HTTPS://TWITTER.COM/PRATHMESHRANAUT
? Prathmesh Ranaut, 2017

Rails is a web application framework written in Ruby that makes programming web apps easier by making assumptions to get developers started quickly. It utilizes conventions like DRY, MVC, and RESTful design. MVC separates business logic from presentation logic into models, views, and controllers. Controllers receive requests, interact with models to retrieve data, and render views to generate responses. Developers can also use Ruby gems to add additional functionality like user authentication, file uploads, and payment processing to their Rails applications.

RESTful APIとしてのRailsとクライアントとしてのJavaScriptNaoto Koshikawa

搁补颈濒蝉のフロントエンド开発を考えるHirata Tomoko

Railsエンジニアがサーバーレスアーキテクチャに手を出したよ - 川崎Ruby会議01Yuta Shimizu

Rails あるあるRyunosuke SATO

搁贰厂罢蹿耻濒开発フロントエント?编（厂笔础?础濒迟闯厂?フレームワーク）K Tsukada

RESTful #とは RailsスタイルからRESTを学ぼうToru Kawamura

RHP like Rapid Home ProvisioningDaniele Massimi

Developing Web Services from Scratch - For DBAs and Database DevelopersRevelation Technologies

WSDL. XSD. SOAP. Namespaces. Port types. If these terms make little sense, this presentation is for you. By the end of this presentation, you will completely understand how to dissect and decipher a web service interface, understand key design patterns, and learn how to develop top-down and bottom-up web services in technologies such as Java and Oracle SOA Suite. Want to know how to expose a PL/SQL package as a web service? This technical presentation, one of my most popular, is intended for DBAs and database developers who want to know what it takes to design and create web services.

Page Performancestk_jj

The document discusses optimizing WordPress page performance. It begins by explaining why performance is important for user experience, conversions, and retention. It then defines key performance metrics like Time to First Byte (TTFB), First Input Delay (FID), Cumulative Layout Shift (CLS), Largest Contentful Paint (LCP), page size, and number of requests. For each metric, it provides recommendations on targets to aim for to improve performance. Additional tips include using efficient themes, images, caching, minification, concatenation, lazy loading, and content delivery networks. The document emphasizes measuring performance and provides several tools and resources for doing so.

ROMA NOVIKOV, BAQ, "Prometheus + grafana based monitoring"Dakiry

This document provides an agenda and summary of a presentation on Prometheus and Grafana based monitoring. The presentation discusses why monitoring is needed, common monitoring tools, methods like USE and RED, and how Prometheus and Grafana can be used together for cloud native monitoring. It also provides an overview of Percona Monitoring and Management (PMM) as a simplified and open source solution for Prometheus, Grafana and MySQL monitoring.

Downsampling your data October 2017InfluxData

This document discusses using continuous queries and retention policies in InfluxDB to downsample and manage the retention of time series data. It provides examples of: - Creating continuous queries to periodically aggregate high resolution data into lower resolution measurements - Creating retention policies to configure how long raw and aggregated data is stored - A case study combining continuous queries and retention policies to downsample 10-second telemetry data to 5-minute and store for different durations

Three in12c: Row Limiting, PL/SQL With SQL and Temporal ValidityTrivadis

RedisConf17 - Amadeus - Redis-Cluster operatorRedis Labs

This document discusses making Redis cluster deployments easier using an operator on OpenShift. It motivates using Redis cluster for session handling and improved performance. It describes prototypes that deployed Redis cluster but lacked features like simple deployment, updating topology, and rolling updates. The document then presents a Redis manager application that automates cluster operations on OpenShift and proposes developing a Redis operator to improve lifecycle management and support multiple clusters. It concludes by discussing open sourcing the Redis manager and developing it further as an operator.

Introduction to Performance Testing & LoadrunnerAisha Mazhar

This document discusses HP LoadRunner, a performance testing tool. It provides an overview of LoadRunner, including what performance testing is, the types of performance testing, limitations of manual testing, LoadRunner components, designing and executing scenarios, and analyzing results. The key points are that LoadRunner automates performance testing by using virtual users to simulate real user loads and measure system behavior, components include VuGen, Controller, and Load Generators, and it allows designing, running, and analyzing load testing scenarios to evaluate system performance.

[RedisConf17] Redis Cluster Operability with Kubernetes and OpenShift - Cedri...Cedric Lamoriniere

Network & Application Performance Monitoring - Troubleshooting HTTP Applicati...PerformanceVision (previously SecurActive)

The document discusses network and application performance monitoring (N/APM) using Performance Vision. It provides an overview of Performance Vision's capabilities such as identifying application bottlenecks, troubleshooting HTTP application performance, and network performance monitoring. The presentation demonstrates how Performance Vision can be used to diagnose various issues like application failures, slow page loads, and login problems. It also highlights Performance Vision's non-intrusive deployment and fast troubleshooting abilities.

More tips and tricks for running containers like a pro - Rancher Online MEetu...Shannon Williams

Demystifying Web PerformanceAtlassian

"Measure first, then optimise" is an oft-cited maxim for developers. But you can't optimise without first understanding what factors influence a given measurement. Only then is it worth your time to start digging around in your code and infrastructure. Join Atlassian engineer Ian Grunert as he outlines the measurements exposed by the browser about the request lifecycle, and the areas of your web application contributing to each section. He'll also show you how the Confluence team uses Bamboo, Bitbucket (including Stash), Confluence, and JIRA to track the impact of performance improvements and reduce the chance of shipping performance regressions.

Oracle BPM Suite Development: Getting StartedRevelation Technologies

The document outlines a presentation about getting started with Oracle BPM Suite development. It discusses downloading and installing the Oracle BPM Suite quick start, an overview of the key concepts in BPM like service types, gateways and sampling points. It also covers capabilities like BPM scripting using Groovy and case management applications. The presentation aims to explain why Oracle BPM is useful for providing greater visibility into business processes and supporting more complex processes and manual steps.

Webinar 2017. Supercharge your analytics with ClickHouse. Vadim TkachenkoAltinity Ltd

ClickHouse is an open-source column-oriented database management system designed for high speed analytics on large data sets. It was developed by Yandex to handle real-time reporting and analytics for their search engine and ad analytics platforms, processing billions of events daily. ClickHouse provides massively parallel processing, distributed query processing, and high reliability. Benchmarks show it outperforms other open-source and commercial analytics databases in speed. It is best suited for analytics workloads like web and app analytics, sensor data, and monitoring where fast queries and scalability to large data volumes are important.

Orchestrating Big Data pipelines @ Fandom - Krystian Mistrzak Thejas MurthyEvention

Fandom is the largest entertainment fan site in the world. With more than 360,000 fan communities and a global audience of over 190 million monthly uniques, we are the fan’s voice in entertainment. Being the largest entertainment site, wikia generates massive volumes of data, which varies from clickstream, user activities, api requests, ad delivery, A/B testing and much more. The big challenge is not just the volume but the orchestration involved in combining various sources of data with various periodicity, volumes. And Making sure the processed data is available for the consumers within the expected time. Thus helping gain the right insights well within the right time. A conscious decision was made to choose the right open source tool to solve the problem of orchestration, after evaluating various tools we decided to use Apache airflow. This presentation will give an overview of comparisons of existing tools and emphasize on why we choose airflow. And how Airflow is being used to create a stable reliable orchestration platform to enable non data engineers to seamlessly access data by democratizing data. We will focus on some tricks and best practises of developing workflows with Airflow and show how we are using some of the features of airflow.

Rails App performance at the limit - Bogdan GusievRuby Meditation

The document discusses optimizing Rails application performance. It begins by examining where Rails may be slow, such as ActiveRecord and Action Controller. The author advocates for effective database schema as the best general optimization technique. Specific optimizations should target problematic parts of the application and may include avoiding ActiveRecord, conditional GET requests, rewriting controllers in Rack, and introducing caching. The goals of optimization are to improve throughput, reduce infrastructure costs, and make targeted improvements where needed.

ShadowReader - Serverless load tests for replaying production trafficYuki Sawa

While load testing has become more accessible, configuring load tests that faithfully recreate production conditions can be difficult -- a good load test must use a set of URLs that is representative of production traffic and achieve request rates that mimic real users. Even performing distributed load tests requires the upkeep of a fleet of servers. ShadowReader aims to solve these problems. It gathers URLs and request rates straight from production logs and replays it using AWS Lambda. Being serverless, it is more efficient cost and performance wise than traditional distributed load tests and in practice has scaled beyond 50,000 requests / minute. At Edmunds, we have been able to utilize these capabilities to solve problems such as Node.js memory leaks that were happening only in production by recreating the same conditions in our QA environment. It’s also being used everyday to generate load for pre-prod canary deployments. This presentation will go over: - How ShadowReader solved a production incident through replaying traffic - Explain ShadowReader's serverless architecture - Show the audience how they can leverage it for their own testing ShadowReader has recently been open sourced on GitHub and is actively seeking suggestions and contributions.

Oracle Database Cloud Service - Provisioning Your First DBaaS InstanceRevelation Technologies

Moving to the cloud is inevitable, yet many haven’t started getting their hands dirty yet. This presentation will walk you through key Oracle Cloud concepts for those who have never seen or used it before, and walk through provisioning a cloud database, which includes storage and infrastructure. We explain how to access it, and discuss various administration concepts. This presentation is perfect for the Oracle DBA who wants to understand how to get started administering an Oracle DBaaS environment, from a technical perspective.

Scaling XGBoost to large scale clusters with fault tolerance and recoveryChen Qin

Monitor your Java application with Prometheus StackWojciech Barczyński

Artificial-Intelligence-in-Cybersecurity.pptxVigneshwarar3

Analysis of Daylighting in Interior Spaces using the Daylight Factor - A Manu...Ignacio J. J. Palma Carazo

More Related Content

Similar to Performance Optimization in Ruby (20)

RHP like Rapid Home ProvisioningDaniele Massimi

Developing Web Services from Scratch - For DBAs and Database DevelopersRevelation Technologies

Page Performancestk_jj

ROMA NOVIKOV, BAQ, "Prometheus + grafana based monitoring"Dakiry

Downsampling your data October 2017InfluxData

Three in12c: Row Limiting, PL/SQL With SQL and Temporal ValidityTrivadis

RedisConf17 - Amadeus - Redis-Cluster operatorRedis Labs

Introduction to Performance Testing & LoadrunnerAisha Mazhar

[RedisConf17] Redis Cluster Operability with Kubernetes and OpenShift - Cedri...Cedric Lamoriniere

Network & Application Performance Monitoring - Troubleshooting HTTP Applicati...PerformanceVision (previously SecurActive)

More tips and tricks for running containers like a pro - Rancher Online MEetu...Shannon Williams

Demystifying Web PerformanceAtlassian

Oracle BPM Suite Development: Getting StartedRevelation Technologies

Webinar 2017. Supercharge your analytics with ClickHouse. Vadim TkachenkoAltinity Ltd

Orchestrating Big Data pipelines @ Fandom - Krystian Mistrzak Thejas MurthyEvention

Rails App performance at the limit - Bogdan GusievRuby Meditation

ShadowReader - Serverless load tests for replaying production trafficYuki Sawa

Oracle Database Cloud Service - Provisioning Your First DBaaS InstanceRevelation Technologies

Scaling XGBoost to large scale clusters with fault tolerance and recoveryChen Qin

Monitor your Java application with Prometheus StackWojciech Barczyński

RHP like Rapid Home ProvisioningDaniele Massimi

Developing Web Services from Scratch - For DBAs and Database DevelopersRevelation Technologies

Page Performancestk_jj

ROMA NOVIKOV, BAQ, "Prometheus + grafana based monitoring"Dakiry

Downsampling your data October 2017InfluxData

Three in12c: Row Limiting, PL/SQL With SQL and Temporal ValidityTrivadis

RedisConf17 - Amadeus - Redis-Cluster operatorRedis Labs

Introduction to Performance Testing & LoadrunnerAisha Mazhar

[RedisConf17] Redis Cluster Operability with Kubernetes and OpenShift - Cedri...Cedric Lamoriniere

Network & Application Performance Monitoring - Troubleshooting HTTP Applicati...PerformanceVision (previously SecurActive)

More tips and tricks for running containers like a pro - Rancher Online MEetu...Shannon Williams

Demystifying Web PerformanceAtlassian

Oracle BPM Suite Development: Getting StartedRevelation Technologies

Webinar 2017. Supercharge your analytics with ClickHouse. Vadim TkachenkoAltinity Ltd

Orchestrating Big Data pipelines @ Fandom - Krystian Mistrzak Thejas MurthyEvention

Rails App performance at the limit - Bogdan GusievRuby Meditation

ShadowReader - Serverless load tests for replaying production trafficYuki Sawa

Oracle Database Cloud Service - Provisioning Your First DBaaS InstanceRevelation Technologies

Scaling XGBoost to large scale clusters with fault tolerance and recoveryChen Qin

Monitor your Java application with Prometheus StackWojciech Barczyński

Recently uploaded (20)

Artificial-Intelligence-in-Cybersecurity.pptxVigneshwarar3

Analysis of Daylighting in Interior Spaces using the Daylight Factor - A Manu...Ignacio J. J. Palma Carazo

Energy Transition Factbook Bloomberg.pdfCarlosdelaFuenteMnde

Unit-03 Cams and Followers in Mechanisms of Machines.pptxKirankumar Jagtap

Water Industry Process Automation & Control Monthly - April 2025Water Industry Process Automation & Control

Welcome to the April 2025 edition of WIPAC Monthly, the magazine brought to you by the LInkedIn Group Water Industry Process Automation & Control. In this month's issue, along with all of the industries news we have a number of great articles for your edification The first article is my annual piece looking behind the storm overflow numbers that are published each year to go into a bit more depth and look at what the numbers are actually saying. The second article is a taster of what people will be seeing at the SWAN Annual Conference next month in Berlin and looks at the use of fibre-optic cable for leak detection and how its a technology we should be using more of The third article, by Rob Stevens, looks at what the options are for the Continuous Water Quality Monitoring that the English Water Companies will be installing over the next year and the need to ensure that we install the right technology from the start. Hope you enjoy the current edition, Oliver

Requirements Engineering for Secure SoftwareDr Sarika Jadhav

Security requirements are often treated as generic lists of features, neglecting system-specific needs and the attacker's perspective. A systematic approach to security requirements engineering is crucial to avoid this problem. Requirements engineering defects can cost 10 to 200 times more to correct once the system is operational. Software development takes place in a dynamic environment, causing requirements to constantly change.

module-4.1-Class notes_R and DD_basket-IV -.pdfritikkumarchaudhury7

Distributed renewable energy in Colombia.OECD2023.pdfSantiagoCardonaGallo

Floating Offshore Wind in the Celtic Seapermagoveu

Floating offshore wind (FLOW) governance arrangements in the Celtic Sea case are changing and innovating in response to different drivers including domestic political priorities (e.g. net-zero, decarbonization, economic growth) and external shocks that emphasize the need for energy security (e.g. the war in Ukraine). To date, the rules of the game that guide floating wind in the UK have evolved organically rather than being designed with intent, which has created policy incoherence and fragmented governance arrangements. Despite this fragmentation, the UK has a well-established offshore wind sector and is positioning itself to become a global leader in floating wind. Marine governance arrangements are in a state of flux as new actors, resources, and rules of the game are being introduced to deliver on this aspiration. However, the absence of a clear roadmap to deliver on ambitious floating wind targets by 2030 creates uncertainty for investors, reduces the likelihood that a new floating wind sector will deliver economic and social value to the UK, and risks further misalignment between climate and nature goals.

????? ??? ???? 2025 (Lok Fitting Catalog 2025)????? / HIFLUX Co., Ltd.

Software security: Security a Software IssueDr Sarika Jadhav

Software is often designed with security as an afterthought, leading to vulnerabilities that can be exploited by attackers. This has become a critical issue as our reliance on software continues to grow. Increasing number and sophistication of attacks (CERT vulnerability reports rising). Software security is the practice of protecting applications from unauthorized access, modification, and destruction. Secure software development practices. Executives (E) Project Managers (M) Technical Leaders (L)

Ktor - Definizioni di Path, Integrazioni, Plugin e build fino al rilascioinfogdgmi

Scalling Rails: The Journey to 200M NotificationsGustavo Araujo

Introduction to Forensic Research Digital ForensicsSaanviMisar

Digital Forensics: Analyzing Cyber Crimes & Investigations This comprehensive guide on Digital Forensics covers key concepts, tools, and methodologies used in investigating cyber crimes. It explores forensic techniques, evidence collection, data recovery, malware analysis, and incident response with real-world applications. Topics Covered: Introduction to Digital Forensics Cybercrime Investigation Process Digital Evidence & Chain of Custody Popular Forensic Tools (Autopsy, EnCase, FTK) Memory & Network Forensics Challenges in Modern Cyber Investigations Ideal for students, cybersecurity professionals, and forensic analysts, this resource provides valuable insights into digital investigations.

Crude-Oil-System for oil and gas industryOkeke Livinus

e-health to improve the effectiveness of the Healthcare systemDr INBAMALAR T M

Mastering Secure Login Mechanisms for React Apps.pdfBrion Mario

English presentation, tests and experiments.pptxSamahEL2

Production Planning & Control and Inventory Management.pptxVirajPasare

Protecting Secrets in Transparent SystemsLucaBarbaro3

Artificial-Intelligence-in-Cybersecurity.pptxVigneshwarar3

Analysis of Daylighting in Interior Spaces using the Daylight Factor - A Manu...Ignacio J. J. Palma Carazo

Energy Transition Factbook Bloomberg.pdfCarlosdelaFuenteMnde

Unit-03 Cams and Followers in Mechanisms of Machines.pptxKirankumar Jagtap

Water Industry Process Automation & Control Monthly - April 2025Water Industry Process Automation & Control

Requirements Engineering for Secure SoftwareDr Sarika Jadhav

module-4.1-Class notes_R and DD_basket-IV -.pdfritikkumarchaudhury7

Distributed renewable energy in Colombia.OECD2023.pdfSantiagoCardonaGallo

Floating Offshore Wind in the Celtic Seapermagoveu

????? ??? ???? 2025 (Lok Fitting Catalog 2025)????? / HIFLUX Co., Ltd.

Software security: Security a Software IssueDr Sarika Jadhav

Ktor - Definizioni di Path, Integrazioni, Plugin e build fino al rilascioinfogdgmi

Scalling Rails: The Journey to 200M NotificationsGustavo Araujo

Introduction to Forensic Research Digital ForensicsSaanviMisar

Crude-Oil-System for oil and gas industryOkeke Livinus

e-health to improve the effectiveness of the Healthcare systemDr INBAMALAR T M

Mastering Secure Login Mechanisms for React Apps.pdfBrion Mario

English presentation, tests and experiments.pptxSamahEL2

Production Planning & Control and Inventory Management.pptxVirajPasare

Protecting Secrets in Transparent SystemsLucaBarbaro3

Performance Optimization in Ruby

1. PERFORMANCE OPTIMIZATION IN RUBYPRATHMESH RANAUT ? Prathmesh Ranaut, 2017

2. ? Prathmesh Ranaut, 2017

3. ? -> " ? Prathmesh Ranaut, 2017

4. WHO AM I > Computer Science Undergrad > Helped ! of ruby apps improve performance ? Prathmesh Ranaut, 2017

5. > Celluloid - Concurrent ruby framework > Improved performance by 300% ? Prathmesh Ranaut, 2017

6. WHAT IS PERFORMANCE OPTIMIZATION? ! ? Prathmesh Ranaut, 2017

7. DOING MOREWITH LESS ? Prathmesh Ranaut, 2017

8. HOW? ? Prathmesh Ranaut, 2017

9. TIP #1 ? Prathmesh Ranaut, 2017

10. ALWAYS HAVE A BENCHMARKING SUITE ? Prathmesh Ranaut, 2017

11. BECHMARK-IPS ? Prathmesh Ranaut, 2017

12. require "benchmark/ips" Benchmark.ips do |ips| ips.report("addition") { 1 + 2 } end ? Prathmesh Ranaut, 2017

13. require "benchmark/ips" Benchmark.ips do |ips| ips.report("addition") { 1 + 2 } end Warming up -------------------------------------- addition 325.015k i/100ms Calculating ------------------------------------- addition 12.544M (± 8.5%) i/s - 62.403M in 5.012765s ? Prathmesh Ranaut, 2017

14. class Worker include Celluloid def hashed(hash, key) hash[key] end end # Worker pool of different sizes pool_10 = Worker.pool(size: 10) pool_100 = Worker.pool(size: 100) hash = {} ENTRIES = 10_000 TESTS = 100_000 # Using a constant key, to derive consistent benchmarking results KEY = 500 # Populate hash array ENTRIES.times do |i| hash[i] = i end Benchmark.bmbm do |ips| puts "Finding the key : #{KEY}" ips.report("pool - 10") do TESTS.times do pool_10.async.hashed(hash, KEY) end end ips.report("pool - 100") do TESTS.times do pool_100.async.hashed(hash, KEY) end end end ? Prathmesh Ranaut, 2017

15. TIP #2 ? Prathmesh Ranaut, 2017

16. UPGRADE RUBY ? Prathmesh Ranaut, 2017

17. RUBY 1.9 > YARV ? Prathmesh Ranaut, 2017

18. RUBY 1.9 > YARV RUBY 2.0 > Copy-on-Write ? Prathmesh Ranaut, 2017

19. RUBY 2.1 > Generation GC RUBY 2.4 > Improves the speed to access an instance variable ? Prathmesh Ranaut, 2017

20. TIP #3 ? Prathmesh Ranaut, 2017

21. TRY ANOTHER IMPLEMENTATION OF RUBY ? Prathmesh Ranaut, 2017

22. RUBY 2.3 VS JRUBY RUBY 2.3 Calculating ------------------------------------- addition 12.544M (± 8.5%) i/s - 62.403M JRUBY Calculating ------------------------------------- addition 22.929M (± 9.3%) i/s - 113.127M ? Prathmesh Ranaut, 2017

23. TIP #4 ? Prathmesh Ranaut, 2017

24. PROFILE YOUR METHODS ? Prathmesh Ranaut, 2017

25. TIP #5 ? Prathmesh Ranaut, 2017

26. FIGURE OUT THE BOTTLENECK I/O OR CPU OR MEMORY ? Prathmesh Ranaut, 2017

27. TIP #6 ? Prathmesh Ranaut, 2017

28. SWITCH TO ELIXIR ? Prathmesh Ranaut, 2017

29. TRY MULTITHREADING ? Prathmesh Ranaut, 2017

30. TOOLS THAT MAKE MY LIFE EASIER ? Prathmesh Ranaut, 2017

31. METHOD PROFILER HTTPS://GITHUB.COM/CHANGE/METHOD_PROFILER ? Prathmesh Ranaut, 2017

32. profiler = MethodProfiler.observe(BazingaClass) # Do stuff with the BazingaClass puts profiler.report ? Prathmesh Ranaut, 2017

33. ? Prathmesh Ranaut, 2017

34. KCACHEGRIND HTTPS://KCACHEGRIND.GITHUB.IO/ ? Prathmesh Ranaut, 2017

35. ? Prathmesh Ranaut, 2017

36. RUBYPROF HTTPS://GITHUB.COM/RUBY-PROF/RUBY-PROF ? Prathmesh Ranaut, 2017

37. require 'ruby-prof' # profile the code result = RubyProf.profile do # ... code to profile ... end # print a graph profile to text printer = RubyProf::GraphPrinter.new(result) printer.print(STDOUT, {}) ? Prathmesh Ranaut, 2017

38. ? Prathmesh Ranaut, 2017

39. ObjectSpace.count_objects ? Prathmesh Ranaut, 2017

40. { :TOTAL=>30165, :FREE=>2216, :T_OBJECT=>928, :T_CLASS=>574, :T_MODULE=>38, :T_FLOAT=>4, :T_STRING=>13182, :T_REGEXP=>77, :T_ARRAY=>2715, :T_HASH=>88, :T_STRUCT=>3, :T_BIGNUM=>2, :T_FILE=>30, :T_DATA=>363, :T_MATCH=>233, :T_COMPLEX=>1, :T_SYMBOL=>6, :T_IMEMO=>4530, :T_NODE=>5138, :T_ICLASS=>37 } ? Prathmesh Ranaut, 2017

41. DTRACE ? Prathmesh Ranaut, 2017

42. THANK YOU & QUESTIONSHELLO@PRATHMESHRANAUT.COM HTTPS://TWITTER.COM/PRATHMESHRANAUT ? Prathmesh Ranaut, 2017

狠狠撸

Performance Optimization in Ruby

Recommended

More Related Content

Similar to Performance Optimization in Ruby (20)

Recently uploaded (20)

Performance Optimization in Ruby