프로그램
SESSION 1 ~
Real-time large data at Twitter
Twitter's architecture deals with "now". Everything must happen in real-time and the real-time constraint is its hardest problem. At steady state, Twitter receives thousands of tweets a second that it needs to deliver to disks, in-memory timelines, email, and mobile devices. Simultaneously, Twitter deals with changes to its graph -- each change affects the way that data flows through the system. This talk delves into the three nouns of twitter: tweets, timelines, users, and usin...
Raffi Krikorian | Twitter
SESSION 2 ~
Apache Kafka: Inside LinkedIn's distributed publish/subscribe messaging system
Richard Park graduated from University of Waterloo in Computer Science. For the past three years, he has been a software engineer in the Search, Network and Analytics group at LinkedIn. His primary focus has been on distributed systems, and has helped grow LinkedIn's Hadoop infrastructure from dozens to thousands of nodes. He has also assisted in integrating Kafka and Voldemort into LinkedIn's data pipeline, and is a primary developer on Azkaban. Richard has previously worked on fraud detection s...
Richard Park | LinkedIn
SESSION 3 ~
Couchbase Server for Speed and Scale with Interactive Applications
Perry Krug is a Solutions Architect at Couchbase working with customers in all capacities to aid in their experiences with Couchbase. He has been with Couchbase for 2 years, and has been working with high performance caching and database systems for over 6. When developing web or mobile applications for today, system architects need to be ready for a few new challenges. Users expect their application data to be available immediately from multiple 'screens'. &nbs...
Perry Krug | CouchBase
SESSION 4 ~
HDFS Architecture:How HDFS is Evolving to Meet New Needs
Aaron T. Myers is a Software Engineer at Cloudera and an Apache Hadoop Committer. Aaron’s work is primarily focused on HDFS and Hadoop Security. Aaron holds both an Sc.B. and Sc.M. in Computer Science from Brown University. HDFS has for several years been a very capable file system for supporting batch-oriented MapReduce workloads. As HDFS's use cases have expanded, so too have it's requirements, including real-time data read/write, high availability, and increased scalability. This talk w...
Aaron T. Myers | Cloudera
SESSION 5 ~
Enhancing the Scalability of Memcached
Rajiv Kapoor is a Principal Engineer with Intel Corporation’s Software and Services group. His area of expertise is application performance analysis and platform architecture. He has been with Intel for 13 years working on analysis of applications and platforms in different segments from clients to servers. Over the past several years he has been focused on analysis and optimization web search engines and cloud applications/workloads leading to definition of future platform feat...
Rajiv Kapoor | Intel
SESSION 6 ~
Heroku PostgreSQL: The Tale of Conceiving and Building a Leading Cloud Database Service
Harold is an engineer at the Heroku Department of Data, who run the largest fleet of Postgres databases in the world. He always had an interest in data management and usage. In his career he has developed hardware based neural networks, devised and implemented algorithms that crunched through decades of health care data, and is currently building systems that handle all aspects of a cloud database service. Running a database service is arguably the toughest job in the cloud operations industry...
Harold Giménez | Heroku