Taming Big Data with Berkeley Data Analytics Stack

One of the most interesting developments over the past decade is the rapid increase in data; we are now deluged by data from on-line services (PBs per day), scientific instruments (PBs per minute), gene sequencing (250GB per person) and many other sources. Researchers and practitioners collect this massive data with one goal in mind: extract "value" through sophisticated exploratory analysis, and use it as the basis to make decisions as varied as personalized treatment and ad targeting. Unfortunately, today's data analytics tools are slow in answering even simple queries, as they typically require to sift through huge amounts of data stored on disk, and are even less suitable for complex computations, such as machine learning algorithms. These limitations leave the potential of extracting value of big data unfulfilled. To address this challenge, we are developing the Berkeley Data Analytics Stack (BDAS), an open-source data analytics stack that provides interactive response times for complex computations on massive data. To achieve this goal, BDAS supports efficient, large-scale in-memory data processing, and allows users and applications to trade between query accuracy, time, and cost. In this talk, I'll present the architecture, challenges, early results, and our experience with developing BDAS. Some BDAS components have already been released: Mesos, a platform for cluster resource management has been deployed by Twitter on 5,000+ servers, while Spark, an in-memory cluster computing frameworks, is already being used by tens of companies and research institutions.