Spark and Tachyon - Big Data Utah Meeting @ IHC

Hello Everyone,

Jim Lohse will be presenting in February about Spark and Tachyon. Details are below. We'll look forward to seeing you there.

Jim Lohse is a student at the University of Utah seeking a degree in Statstics with a minor in CS, and an emphasis on Machine Learning and Data Mining. He currently has a student programmer part-time job at Huntsman Cancer Institute. He will present a use case for Spark and Tachyon as used in the research project where he works with a lot of very smart.

Jim will also present use cases for Baidu presented in the SF Tachyon group last year. In addition he will demonstrate basic installation and configuration of Tachyon.

Oh, what's is Tachyon? It's a memory-based storage system that, for example, can sit between Hadoop and Spark. From the FAQ at tachyon-project.org (http://tachyon-project.org/): Tachyon (http://tachyon-project.org/) is an open source, memory-centric, distributed, and reliable storage system enabling data sharing across clusters at memory speed. Tachyon uses UnderFS (Underlying File System) as its persistent storage system. It currently supports Amazon S3, HDFS, GlusterFS, and there are more options coming.