Papers have been received from organizations including Apple, Constant Contact, Netflix and Pager Duty. Stay tuned for the confirmed Summit 2015 agenda to be announced this Summer. Register your seat for sessions today!
Speakers from Cassandra Summit 2014
Leena Borle, Sr. Technology Lead, Apple
I am Sr. Technology Lead in the team that handles high volume data. I mainly work on NoSql platform of our system. I have been working on Cassandra since its version 0.7 was released. Our systems are highly available, fault tolerant. For that Cassandra makes a good choice to add it to several parts of our software system.
Cassandra as Caching Technology
Typically when we think of Cassandra, we think of it as a data store for write heavy transaction traffic. But I would like to present a use case where we used Cassandra for Caching. Caching means fast keyed data access, usually for read heavy usages, and static data with small change in velocity. Last year we had a requirement for providing distributed caching solution, which also needed to support following criteria: Highly durable (cache is not static) 2. Consistent 3. Cross datacenter aware 4. Highly available across datacenters 5. Clustering (replication of 3) 6. Complex selection criteria (required a little more than distributed hash tables). Based on above requirements, and our internal evaluation process of other caching technologies, Cassandra was a clear winner. The way we architected our solution was every cache write (put, putIfAbsent(), getandTouch()) gets written across datacenters simultaneously. These calls fail if data is not written in all datacenters. Reads are all local to all datacenters. In this version, while writing data across datacenters, we enabled Cassandra’s Cross-DC setups as well as using REST calls to write same data to Cassandra in each DC. This ensures that data will still be available if one way of writes fails. Cassandra tuning: Our cache data expires in 5 to 15 minutes. We wanted to keep as many keys as possible in the key-cache until TTL, so we adjusted the key-cache size parmeter in YAML file. Frequently TTL ed data means lots of tombstones. We change gc_grace_seconds to 0, so that expired data will get removed as early as possible, since we don’t perform any deletes on those rows from the application layer. I hope this information will be helpful for those who are evaluating caching solutions and is interested in using Cassandra for their needs.
Rick Branson, Software Engineer, Instagram
Rick is a Software Engineer at Instagram where he spends his days building scalable systems to deal with their meteoric growth rates.
Cassandra at Instagram 2014
As Instagram has scaled to over 200 million users, so has our use of Cassandra. We’ve built new features and rebuilt old on Cassandra, and it’s become an extremely mission-critical foundation of our production infrastructure. Rick will deliver a refresh of our use cases and go deep on the technical challenges we faced during our expansion.
Jason Brown, Senior Software Engineer, Apple
Jason Brown is a Senior Software Engineer at Apple. Prior to that, he worked at Netflix as a Senior Software Engineer of web services and data architecture. Jason is also an Apache Cassandra Committer. He holds a Master’s degree in Music Composition and is searching for time to write a second string quartet.
Cassandra Internals: Understanding Gossip
In this talk we’ll dive into how Cassandra nodes discover and communicate with each other, and share global state information via gossip. As the gossip subsystem seems shrouded in mystery to many folks, we’ll peel back the layers and learn how it powers the underbelly of Cassandra.
Dan Cundiff, Technical Architect Consultant, Target Corporation
Dan is a technical architect consultant in Technology Services (TTS). Since 2005 Dan has been doing development in various areas of TTS: data security, standing up internal social and knowledge management platforms, web development, and more. Today he works on a creating RESTful APIs that are used in and outside of the enterprise across a wide range of devices, applications, and business partners. Dan enjoys automating-all-the-things and likes to exchange pro tips on continuous integration, continuous deployment, devops, and infrastructure as code techniques.
Apache Cassandra at Target: Pioneering NoSQL in a Big Enterprise
This presentation will cover the problems we needed to solve, the journey we took to get there, and the lessons we learned along the way. We’ll cover the technical and non-technical aspects of this story.
Tim Czerniak, Software Engineer, Demonware
A graduate of Trinity College, Dublin, Tim has worked with Demonware for the last 6 years, primarily on a team who design and maintain the core software platforms on which Demonware’s services run.
Deploying Cassandra for Call of Duty
This presentation covers the eight-month evaluation process we underwent to migrate some of Call of Duty’s core services from MySQL to Cassandra. We will outline our requirements, the process we followed for the evaluation, decisions we made around our schema, configuration and hardware, and some issues we encountered.
Alexander Filipchik, Staff Software Engineer, Sony Network Entertainment
Alex works as a Staff engineer for Sony Network Entertainment and spends his time designing and scaling PS4 commerce ecosystem. We made a decision to use Cassandra for PS4 5 month before the launch and then proved that 5 month is enough time to go from the design to the full production with a horizontally scalable architecture.
Sony PS4 and Cassandra. Lessons Learned.
Christos Kalantzis, Engineering Manager – Cloud Persistence Engineering, Netflix
Christos has 15+ years of Data experience. He started his career as an Operational DBA in MS SQL Server, DB2 and MySQL for companies such as Matrox, IBM, InterTrade and HighTail. He has helped design and scale applications for petabytes of data, as a Data Architect, for HighTail and Chegg. He has managed engineering teams at HighTail were he and his team created a Cloud Based File System. He is currently managing the Cloud Database Engineering Team at Netflix. His team is developing new data storage solutions for Netflix Engineers, patching and evolving Apache Cassandra, helping other engineering teams with NoSQL designs and operating thousands of Cassandra nodes across hundreds of Cassandra Clusters.
Cassandra @ Netflix: Building a House of Cards on a Solid Foundation
Cassandra is the data storage and global replication foundation for the Cloud Native architecture that runs Netflix streaming for 48+ million members in 40+ countries. In this talk we will show the fundamental role Cassandra has played in our on going global roll-out. We’ve been busy in 2013-2014 experimenting with new use cases with Cassandra, which we will share. We’ve continued our OSS commitment by directly and indirectly contributing to the Apache Cassandra Project and our own NetflixOSS tools: Astyanax, Aegisthus and Priam.
Sankalp Kohli, Software Engineer, Apple
Sankalp Kohli has been working in the Infrastructure Team at Apple for the past 3 years. Currently he is focusing on C* and has done many contributions through various JIRAs. Prior to Apple, he worked at Amazon Web Services on Big data.
Apache Cassandra at Apple for Massive Scale
Cassandra is a critical part of infrastructure that powers various web services at Apple. This session will cover how Apple scaled Cassandra to be used at a massive scale. This talk will touch real world problems we encountered in these areas: AntiEntropy (repair), Streaming and Light weight transactions. This session will also cover recovering Cassandra from a data center failure.
Richard Low, Software Engineer, Apple
Richard has worked with Cassandra for 4 years and is currently working full time on Cassandra in the infrastructure team at Apple. Previously he has worked on analytics and storage products at SwiftKey and Acunu.
Cassandra Doctor at Apple
Everyday at Apple we get deep inside the internals of Cassandra to achieve the best performance and reliability and to troubleshoot issues that come up. In this talk, I will show what tools and techniques we use guided by real-world examples. There’ll be tips for helping with your own clusters as well as practical details on internals.
Chris McEniry, Systems Architect, Sony Network Entertainment
Chris “Mac” McEniry is a practicing sysadmin responsible for running a large ecommerce and gaming service. He’s been working and developing in an operational capacity for 15 years. In his free time, he builds tools and thinks about efficiency.
Common Cassandra Performance Patterns Seen Through Histograms
Apache Cassandra has several built-in granular metrics which let you peer into what actual consumers are seeing, with respect to overall performance. These metrics give you a better insight into specific aspects inside the cluster than the Cassandra clients, which tend to give you overall coarse numbers and are limited in scope. It’s critical for any Cassandra administrator or developer to read and interpret these histograms, to find areas of improvement that will increase overall performance. Luckily, many repeating aspects tend to arise which can lead to a quick resolution. This talk will go over these metrics and what they mean, the situations which can arise to show the common shapes in them, and some tools to help visualize these patterns for quicker recognition.
Seán O Sullivan, Service Reliability Engineer, Demonware
Seán is a service reliability engineer in Demonware for the past three years.
Deploying Cassandra for Call of Duty
This presentation covers the eight month evaluation process we underwent to migrate some of Call of Duty’s core services from MySQL to Cassandra. We will outline our requirements, the process we followed for the evaluation, decisions we made around our schema, configuration and hardware, and some issues we encountered.
Puneet Oberai, Senior Software Engineer, Netflix
Senior Engineer at Netflix
Astyanax: To Be or Not To Be
In this session, we’ll cover a quick introduction to the Astyanax Java client driver, powerful features, comparison to Java Driver and what to do with CQL3.
Dustin Pham, Staff Software Engineer, Sony Network Entertainment
Dustin is a Staff engineer for Sony Network Entertainment and has been one of the main decision makers in using Cassandra for the PS4 ecosystem. He continues to scale out and manage Cassandra in production to meet the demands of the growing playstation user base.
Sony PS4 and Cassandra. Lessons Learned.
Feng Qu, Principal DBA, eBay, Inc.
Starting with Oracle 5, Feng has worked with RDBMS in past 20 years at various high tech companies, including DoubleClick, Yahoo, Intuit and eBay. In recent years, he likes NoSQL technology and has quickly become a huge Cassandra fan while working at eBay. Now he is responsible for eBay/Paypal Cassandra projects from design to deployment.
Apache Cassandra Best Practices at Ebay
Cassandra has been adopted widely at eBay in recent years and used by many end-user facing applications. I will introduce best practices we have built over the time around system design, capacity planning, deployment automation, monitoring integration, performance analysis and troubleshooting. I will also share our experience working with DataStax support to provide a highly available, highly scalable data store fitting into eBay infrastructure.
Anthony Vigil, Software Development Manager, Disney
Software Engineering Manager at The Walt Disney Company
Oauth 2.0 Implementation Using Apache Cassandra at Disney
We have a service which is an Oauth 2.0 implementation at Disney. Initially this service was built using MySQL as a datastore. We migrated the backend to use Cassandra to improve performance, scalability, latency, and availability. Along the way we learned a lot and have been impressed by some things that are much easier with Cassandra than a traditional datastore (such as multi-data center replication and availability).
Gary Stewart, Advisory IT Specialist, ING
Gary is a Dev Engineer at ING with over 14 years of experience, mainly focused around Integration, BPM(Business Process Management) and RDBMS in various industries. In the past year he has really submerged into the NoSQL World to make the bank even more resilient and scalable without trading consistency, “Not an easy switch but definitely worth it!”. Gary contributes to the C* community by co-organizing, presenting and hosting for the Netherlands Cassandra Users meetup group. Within ING, Gary is a trainer, pioneer and evangelist for NoSQL and he is contributing to ING’s engineering culture through learning, sharing and doing.
Apache Cassandra at ING: Testing the Waters
Many financial institutions have discovered NoSQL databases. Some of them are using Cassandra-like technologies in corners of their application landscape to learn, and perhaps as a means to lower the expenses for traditional database suppliers. So did ING. ING is now in production with Cassandra to increase scalability and availability for our channels applications.