Track: Twitter Tech Day (Sponsored)


Day of week:

A special treat for QCon. A technical track with six presentations that dives into the full Twitter stack. Questions on how Twitter provisions hardware, leverages Manhattan (Twitter's distributed key-value store), deploys the JVM, handles Tweet fanout, and Microservices/reliability directly from the SE's building the stack.

Track Host:
Tony Printezis
JVM/GC Engineer @Twitter, JCP Executive Committee, & Previous G1GC Tech Lead
Tony Printezis is a Staff Software Engineer at Twitter and a member of the VM Team in the Infrastructure Organization. He has over 15 years of virtual machine implementation experience with special focus on memory management. Most of his projects have involved improving the performance, scalability, responsiveness, parallelism, concurrency, monitoring, and visualization of garbage collectors. He was one of the designers of (and tech lead of) the G1 GC and the original implementer of the CMS GC. Before Twitter, Tony worked at Adobe, the Java organization at Oracle and Sun Microsystems, and Sun Microsystems Laboratories. He holds a PhD and a BSc(Hons) in Computing Science, both from the University of Glasgow in Scotland.
10:35am - 11:25am

by Nik Johnson
Provisioning Engineering SE @Twitter

by Matt Singer
Staff Hardware Engineer @Twitter

Twitter operates hundreds of thousands of servers. At that scale, repairs and reallocations are frequent and each requires a re-installation of the system. The Provisioning Engineering system at Twitter, called Wilson, is designed to handle every part of a server's lifecycle that does not require a person with a screwdriver. From bios updates to RAID controller configuration and OS installation, the Wilson stack does it all. Combined with our machine metadata service Audubon, Provisioning...

11:50am - 12:40pm

by Boaz Avital
Tech Lead of Manhattan Team @Twitter

Distributed databases are complex systems with multiple dependencies that must be maintained in order to conserve correctness. Manhattan, Twitter's primary distributed key-valve store, has many differing capabilities that require careful and subtly different orchestration while operating on the cluster in order to deploy new versions or add and remove nodes. This talk will touch on the evolution of Manhattan's automated cluster management service, cover some of...

1:40pm - 2:30pm

by John Coomes
Staff Engineer, JVM Team @Twitter

by Ian Downes
Technical Manager Aurora / Mesos Team @Twitter

Twitter services have to be able to handle the flood of requests, responses and actions required to process hundreds of millions of tweets every day.  This talk will cover how we schedule and manage services in our data centers to ensure reliablity and performance, as well as the challenges (and some solutions!)  encountered in getting the JVM and JDK to run our scala- and java-based services to meet stringent response-time and resource requirements.

2:55pm - 3:45pm

by Megan Kanne
Observability Software Engineer @Twitter

Twitter’s Observability team provides Twitter developers a monitoring infrastructure including real time dashboards and alerting for their services. Twitter has grown orders of magnitude since it debuted at SXSW. The monitoring infrastructure has followed suit, ingesting orders of magnitude more metrics, from millions to billions.

As the company grew, so did the challenges of providing an always available alerting system....

4:10pm - 5:00pm

by Moses Nakamura
Core Systems Libraries Software Engineer @Twitter

We choose microservice architectures for many different reasons, including, often, improving reliability. However, there is a dark side. Modular systems have many benefits, but as you add more components, each of which can fail, it makes the overall system more fragile, because there are new ways in which it can fail that previously didn't exist. From its inception, Finagle was designed as a toolkit for reliable systems. We'll discuss understanding the...

5:25pm - 6:15pm

by Mike Cvet
Tech Lead for Timelines Infrastructure Team / Sr Staff Software Engineer @Twitter

Twitter's massive write-time and nearly-instantaneous Tweet fanout system is well-known. This talk will discuss the architecture and challenges involved in the shift towards read-time determination and collection of Tweets for users, incorporating disparate content relevance and recommendation systems. With additional layers of software business logic composed into the core Twitter experience, this has necessitated the adoption of new contracts between teams and...



Monday Nov 7

Tuesday Nov 8

Wednesday Nov 9