The past few years have seen a major change in computing systems, as growing data volumes and stalling processor speeds require more and more applications to scale out to clusters. Today, a myriad data sources, from the Internet to business operations to scientific instruments, produce large and valuable data streams. However, the processing capabilities of single machines have not kept up with the size of data. As a result, organizations increasingly need to scale out their computations over clusters. At the same time, the speed and sophistication required of data processing have grown. In addition to simple queries, complex algorithms like machine learning and graph analysis are becoming common. And in addition to batch processing, streaming analysis of real-time data is required to let organizations take timely action. Future computing platforms will need to not only scale out traditional workloads, but support these new applications too. This book, a revised version of the 2014 ACM Dissertation Award winning dissertation, proposes an architecture for cluster computing systems that can tackle emerging data processing workloads at scale. Whereas early cluster computing systems, like MapReduce, handled batch processing, our architecture also enables streaming and interactive queries, while keeping MapReduce's scalability and fault tolerance. And whereas most deployed systems only support simple one-pass computations (e.g., SQL queries), ours also extends to the multi-pass algorithms required for complex analytics like machine learning. Finally, unlike the specialized systems proposed for some of these workloads, our architecture allows these computations to be combined, enabling rich new applications that intermix, for example, streaming and batch processing. We achieve these results through a simple extension to MapReduce that adds primitives for data sharing, called Resilient Distributed Datasets (RDDs). We show that this is enough to capture a wide range of workloads. We implement RDDs in the open source Spark system, which we evaluate using synthetic and real workloads. Spark matches or exceeds the performance of specialized systems in many domains, while offering stronger fault tolerance properties and allowing these workloads to be combined. Finally, we examine the generality of RDDs from both a theoretical modeling perspective and a systems perspective. This version of the dissertation makes corrections throughout the text and adds a new section on the evolution of Apache Spark in industry since 2014. In addition, editing, formatting, and links for the references have been added.

Learn how to use, deploy, and maintain Apache Spark with this comprehensive guide, written by the creators of the open-source cluster-computing framework. With an emphasis on improvements and new features in Spark 2.0, authors Bill Chambers and Matei Zaharia break down Spark topics into distinct sections, each with unique goals. You'll explore the basic operations and common functions of Spark's structured APIs, as well as Structured Streaming, a new high-level API for building end-to-end streaming applications. Developers and system administrators will learn the fundamentals of monitoring, tuning, and debugging Spark, and explore machine learning techniques and scenarios for employing MLlib, Spark's scalable machine-learning library. Get a gentle overview of big data and Spark Learn about DataFrames, SQL, and Datasets-Spark's core APIs-through worked examples Dive into Spark's low-level APIs, RDDs, and execution of SQL and DataFrames Understand how Spark runs on a cluster Debug, monitor, and tune Spark clusters and applications Learn the power of Structured Streaming, Spark's stream-processing engine Learn how you can apply MLlib to a variety of problems, including classification or recommendation

Train, test, run, track, store, tune, deploy, and explain provenance-aware deep learning models and pipelines at scale with reproducibility using MLflow Key Features Focus on deep learning models and MLflow to develop practical business AI solutions at scale Ship deep learning pipelines from experimentation to production with provenance tracking Learn to train, run, tune and deploy deep learning pipelines with explainability and reproducibility Book DescriptionThe book starts with an overview of the deep learning (DL) life cycle and the emerging Machine Learning Ops (MLOps) field, providing a clear picture of the four pillars of deep learning: data, model, code, and explainability and the role of MLflow in these areas. From there onward, it guides you step by step in understanding the concept of MLflow experiments and usage patterns, using MLflow as a unified framework to track DL data, code and pipelines, models, parameters, and metrics at scale. You'll also tackle running DL pipelines in a distributed execution environment with reproducibility and provenance tracking, and tuning DL models through hyperparameter optimization (HPO) with Ray Tune, Optuna, and HyperBand. As you progress, you'll learn how to build a multi-step DL inference pipeline with preprocessing and postprocessing steps, deploy a DL inference pipeline for production using Ray Serve and AWS SageMaker, and finally create a DL explanation as a service (EaaS) using the popular Shapley Additive Explanations (SHAP) toolbox. By the end of this book, you'll have built the foundation and gained the hands-on experience you need to develop a DL pipeline solution from initial offline experimentation to final deployment and production, all within a reproducible and open source framework. What you will learn Understand MLOps and deep learning life cycle development Track deep learning models, code, data, parameters, and metrics Build, deploy, and run deep learning model pipelines anywhere Run hyperparameter optimization at scale to tune deep learning models Build production-grade multi-step deep learning inference pipelines Implement scalable deep learning explainability as a service Deploy deep learning batch and streaming inference services Ship practical NLP solutions from experimentation to production Who this book is forThis book is for machine learning practitioners including data scientists, data engineers, ML engineers, and scientists who want to build scalable full life cycle deep learning pipelines with reproducibility and provenance tracking using MLflow. A basic understanding of data science and machine learning is necessary to grasp the concepts presented in this book.

The past few years have seen a major change in computing systems, as growing data volumes and stalling processor speeds require more and more applications to scale out to clusters. Today, a myriad data sources, from the Internet to business operations to scientific instruments, produce large and valuable data streams. However, the processing capabilities of single machines have not kept up with the size of data. As a result, organizations increasingly need to scale out their computations over clusters. At the same time, the speed and sophistication required of data processing have grown. In addition to simple queries, complex algorithms like machine learning and graph analysis are becoming common. And in addition to batch processing, streaming analysis of real-time data is required to let organizations take timely action. Future computing platforms will need to not only scale out traditional workloads, but support these new applications too. This book, a revised version of the 2014 ACM Dissertation Award winning dissertation, proposes an architecture for cluster computing systems that can tackle emerging data processing workloads at scale. Whereas early cluster computing systems, like MapReduce, handled batch processing, our architecture also enables streaming and interactive queries, while keeping MapReduce's scalability and fault tolerance. And whereas most deployed systems only support simple one-pass computations (e.g., SQL queries), ours also extends to the multi-pass algorithms required for complex analytics like machine learning. Finally, unlike the specialized systems proposed for some of these workloads, our architecture allows these computations to be combined, enabling rich new applications that intermix, for example, streaming and batch processing. We achieve these results through a simple extension to MapReduce that adds primitives for data sharing, called Resilient Distributed Datasets (RDDs). We show that this is enough to capture a wide range of workloads. We implement RDDs in the open source Spark system, which we evaluate using synthetic and real workloads. Spark matches or exceeds the performance of specialized systems in many domains, while offering stronger fault tolerance properties and allowing these workloads to be combined. Finally, we examine the generality of RDDs from both a theoretical modeling perspective and a systems perspective. This version of the dissertation makes corrections throughout the text and adds a new section on the evolution of Apache Spark in industry since 2014. In addition, editing, formatting, and links for the references have been added.