Inside @Scale 2015

Visit the post for more.

Meghan Marquez
4 min readintermediate
--
View Original

Overview

The @Scale Conference 2015 brought together 1,400 engineers from over 400 companies to share best practices for building scalable systems and applications. Key discussions included mobile development, data processing, and development tools, featuring insights from industry leaders like Facebook, Twitter, and Microsoft.

What You'll Learn

1

How to leverage React Native for mobile app development

2

Why transitioning to a platform-first architecture can enhance mobile development

3

How to implement stream processing at scale using Heron

4

When to use Spark and Cassandra for big data analytics

5

How cloud tools can transform software development practices

Key Questions Answered

What are the benefits of using React Native for mobile app development?
React Native simplifies mobile app development by allowing developers to use the same codebase for both iOS and Android platforms. This leads to faster development cycles and easier maintenance, as seen with its adoption since its introduction in April.
How does Heron improve upon Storm for stream processing?
Heron replaces Storm at Twitter by addressing its challenges, particularly in efficiency, latency, and throughput. Karthik highlighted how Heron's design enhances performance, making it a more suitable choice for real-time data processing.
What architecture did Pinterest adopt for batch-generated data sets?
Pinterest introduced Terrapin, an open-source serving system designed to localize data for improved performance. This architecture helps avoid issues faced with Hadoop on HBase, particularly when handling large data sets.
What insights did Google share about their software development practices?
Google discussed their continuous delivery approach and the role of DevOps in efficient feature launches. They also highlighted tools like Cloud Debugger and Bazel that enhance their engineering culture and productivity.

Key Statistics & Figures

Number of engineers attending the conference
1,400
This figure highlights the scale of participation at the @Scale Conference 2015.
Number of companies represented
400
The diverse representation underscores the collaborative spirit of the engineering community at the event.
Number of talks given by companies
more than a dozen
This indicates the wealth of knowledge shared during the conference.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

Frontend
React Native
Used for mobile app development to streamline the process across platforms.
Data Processing
Heron
Stream processing technology that replaced Storm at Twitter.
Data Processing
Spark
Used for big data analytics solutions at Microsoft.
Database
Cassandra
Utilized alongside Spark for managing large datasets.
Data Serving
Terrapin
Open-sourced by Pinterest for serving batch-generated data sets.

Key Actionable Insights

1
Adopting React Native can significantly streamline your mobile app development process.
By using React Native, developers can maintain a single codebase for both iOS and Android, which reduces development time and complexity.
2
Transitioning to a platform-first architecture can lead to more efficient mobile development.
This approach allows for the creation of reusable components, which can enhance the scalability and maintainability of mobile applications.
3
Implementing Heron for stream processing can lead to better performance metrics compared to Storm.
Heron’s design focuses on efficiency and low latency, making it a compelling choice for organizations dealing with real-time data.
4
Utilizing Spark and Cassandra can provide robust solutions for big data analytics.
These technologies help manage large datasets effectively, allowing for scalable analytics solutions that can adapt to growing data needs.

Common Pitfalls

1
Failing to adopt a platform-first architecture can lead to inefficiencies in mobile development.
Without this approach, teams may struggle with maintaining multiple codebases, leading to increased development time and potential inconsistencies.