Teaching Apache Spark Clusters to Manage Their Workers Elastically: Spark Sum...Spark Summit
Devops engineers have applied a great deal of creativity and energy to invent tools that automate infrastructure management, in the service of deploying capable and functional applications. For data-driven applications running on Apache Spark, the details of instantiating and managing the backing Spark cluster can be a distraction from focusing on the application logic. In the spirit of devops, automating Spark cluster management tasks allows engineers to focus their attention on application code that provides value to end-users.
Using Openshift Origin as a laboratory, we implemented a platform where Apache Spark applications create their own clusters and then dynamically manage their own scale via host-platform APIs. This makes it possible to launch a fully elastic Spark application with little more than the click of a button.
We will present a live demo of turn-key deployment for elastic Apache Spark applications, and share what we’ve learned about developing Spark applications that manage their own resources dynamically with platform APIs.
The audience for this talk will be anyone looking for ways to streamline their Apache Spark cluster management, reduce the workload for Spark application deployment, or create self-scaling elastic applications. Attendees can expect to learn about leveraging APIs in the Kubernetes ecosystem that enable application deployments to manipulate their own scale elastically.
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...Databricks
We’ve all heard that AI is going to become as ubiquitous in the enterprise as the telephone, but what does that mean exactly?
Everyone in IBM has a telephone; and everyone knows how to use her telephone; and yet IBM isn’t a phone company. How do we bring AI to the same standard of ubiquity — where everyone in a company has access to AI and knows how to use AI; and yet the company is not an AI company?
In this talk, we’ll break down the challenges a domain expert faces today in applying AI to real-world problems. We’ll talk about the challenges that a domain expert needs to overcome in order to go from “I know a model of this type exists” to “I can tell an application developer how to apply this model to my domain.”
We’ll conclude the talk with a live demo that show cases how a domain expert can cut through the five stages of model deployment in minutes instead of days using IBM and other open source tools.
David Kale and Ruben Fizsel from Skymind talk about deep learning for the JVM and enterprise using deeplearning4j (DL4J). Deep learning (nouveau neural nets) have sparked a renaissance in empirical machine learning with breakthroughs in computer vision, speech recognition, and natural language processing. However, many popular deep learning frameworks are targeted to researchers and poorly suited to enterprise settings that use Java-centric big data ecosystems. DL4J bridges the gap, bringing high performance numerical linear algebra libraries and state-of-the-art deep learning functionality to the JVM.
Bringing HPC Algorithms to Big Data Platforms: Spark Summit East talk by Niko...Spark Summit
The talk will present a MPI-based extension of the Spark platform developed in the context of light source facilities. The background and rationale of this extension are described in the attached paper “Bringing the HPC reconstruction algorithms to Big Data platforms”[1], which has been presented at New York Scientific Data Summit (NYSDS), August 14-17, 2016 (talk: https://www.bnl.gov/nysds16/files/pdf/talks/NYSDS16%20Malitsky.pdf) Specifically, the paper highlighted a gap between two modern driving forces of the scientific discovery process: HPC and Big Data technologies. As a result, it proposed to extend the Spark platform with inter-worker communication for supporting scientific-oriented parallel applications. The approach was illustrated in the context of the Spark-based deployment of the SHARP MPI/GPU ptychographic solver. Aside from its practical value, this application represents a reference use case that captures the major technical aspects of other reconstruction tasks. In the NYSDS’16 paper, the implemented approach followed the CaffeOnSpark RDMA peer-to-peer model and augmented it with the RDMA address exchange server. By the Spark Summit, we plan to further advance this direction with the Spark-MPI generic solution based on the Hydra process management framework for supporting two major MPI implementations, MPICH and MVAPICH.
Realizing the promise of portable data processing with Apache BeamDataWorks Summit
The world of big data involves an ever changing field of players. Much as SQL stands as a lingua franca for declarative data analysis, Apache Beam aims to provide a portable standard for expressing robust, out-of-order data processing pipelines in a variety of languages across a variety of platforms. In a way, Apache Beam is a glue that can connect the Big Data ecosystem together; it enables users to "run-anything-anywhere".
This talk will briefly cover the capabilities of the Beam model for data processing, as well as the current state of the Beam ecosystem. We'll discuss Beam architecture and dive into the portability layer. We'll offer a technical analysis of the Beam's powerful primitive operations that enable true and reliable portability across diverse environments. Finally, we'll demonstrate a complex pipeline running on multiple runners in multiple deployment scenarios (e.g. Apache Spark on Amazon Web Services, Apache Flink on Google Cloud, Apache Apex on-premise), and give a glimpse at some of the challenges Beam aims to address in the future.
Teaching Apache Spark Clusters to Manage Their Workers Elastically: Spark Sum...Spark Summit
Devops engineers have applied a great deal of creativity and energy to invent tools that automate infrastructure management, in the service of deploying capable and functional applications. For data-driven applications running on Apache Spark, the details of instantiating and managing the backing Spark cluster can be a distraction from focusing on the application logic. In the spirit of devops, automating Spark cluster management tasks allows engineers to focus their attention on application code that provides value to end-users.
Using Openshift Origin as a laboratory, we implemented a platform where Apache Spark applications create their own clusters and then dynamically manage their own scale via host-platform APIs. This makes it possible to launch a fully elastic Spark application with little more than the click of a button.
We will present a live demo of turn-key deployment for elastic Apache Spark applications, and share what we’ve learned about developing Spark applications that manage their own resources dynamically with platform APIs.
The audience for this talk will be anyone looking for ways to streamline their Apache Spark cluster management, reduce the workload for Spark application deployment, or create self-scaling elastic applications. Attendees can expect to learn about leveraging APIs in the Kubernetes ecosystem that enable application deployments to manipulate their own scale elastically.
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...Databricks
We’ve all heard that AI is going to become as ubiquitous in the enterprise as the telephone, but what does that mean exactly?
Everyone in IBM has a telephone; and everyone knows how to use her telephone; and yet IBM isn’t a phone company. How do we bring AI to the same standard of ubiquity — where everyone in a company has access to AI and knows how to use AI; and yet the company is not an AI company?
In this talk, we’ll break down the challenges a domain expert faces today in applying AI to real-world problems. We’ll talk about the challenges that a domain expert needs to overcome in order to go from “I know a model of this type exists” to “I can tell an application developer how to apply this model to my domain.”
We’ll conclude the talk with a live demo that show cases how a domain expert can cut through the five stages of model deployment in minutes instead of days using IBM and other open source tools.
David Kale and Ruben Fizsel from Skymind talk about deep learning for the JVM and enterprise using deeplearning4j (DL4J). Deep learning (nouveau neural nets) have sparked a renaissance in empirical machine learning with breakthroughs in computer vision, speech recognition, and natural language processing. However, many popular deep learning frameworks are targeted to researchers and poorly suited to enterprise settings that use Java-centric big data ecosystems. DL4J bridges the gap, bringing high performance numerical linear algebra libraries and state-of-the-art deep learning functionality to the JVM.
Bringing HPC Algorithms to Big Data Platforms: Spark Summit East talk by Niko...Spark Summit
The talk will present a MPI-based extension of the Spark platform developed in the context of light source facilities. The background and rationale of this extension are described in the attached paper “Bringing the HPC reconstruction algorithms to Big Data platforms”[1], which has been presented at New York Scientific Data Summit (NYSDS), August 14-17, 2016 (talk: https://www.bnl.gov/nysds16/files/pdf/talks/NYSDS16%20Malitsky.pdf) Specifically, the paper highlighted a gap between two modern driving forces of the scientific discovery process: HPC and Big Data technologies. As a result, it proposed to extend the Spark platform with inter-worker communication for supporting scientific-oriented parallel applications. The approach was illustrated in the context of the Spark-based deployment of the SHARP MPI/GPU ptychographic solver. Aside from its practical value, this application represents a reference use case that captures the major technical aspects of other reconstruction tasks. In the NYSDS’16 paper, the implemented approach followed the CaffeOnSpark RDMA peer-to-peer model and augmented it with the RDMA address exchange server. By the Spark Summit, we plan to further advance this direction with the Spark-MPI generic solution based on the Hydra process management framework for supporting two major MPI implementations, MPICH and MVAPICH.
Realizing the promise of portable data processing with Apache BeamDataWorks Summit
The world of big data involves an ever changing field of players. Much as SQL stands as a lingua franca for declarative data analysis, Apache Beam aims to provide a portable standard for expressing robust, out-of-order data processing pipelines in a variety of languages across a variety of platforms. In a way, Apache Beam is a glue that can connect the Big Data ecosystem together; it enables users to "run-anything-anywhere".
This talk will briefly cover the capabilities of the Beam model for data processing, as well as the current state of the Beam ecosystem. We'll discuss Beam architecture and dive into the portability layer. We'll offer a technical analysis of the Beam's powerful primitive operations that enable true and reliable portability across diverse environments. Finally, we'll demonstrate a complex pipeline running on multiple runners in multiple deployment scenarios (e.g. Apache Spark on Amazon Web Services, Apache Flink on Google Cloud, Apache Apex on-premise), and give a glimpse at some of the challenges Beam aims to address in the future.
Using Pluggable Apache Spark SQL Filters to Help GridPocket Users Keep Up wit...Spark Summit
Analyzing and comparing your energy consumption with that of other consumers provides healthy peer pressure and useful insight leading to energy conservation and impacting the bottom line. We helped GridPocket (http://www.gridpocket.com/), a smart grid company developing energy management applications for electricity water and gas utilities, implement high scale anonymized energy comparison queries with an order of magnitude lower cost and higher performance than was previously possible. IoT use cases like that of GridPocket are swamping our planet with data, and drive demand for analytics on extremely scalable and low cost storage. Enter Spark SQL over Object Storage: highly scalable and low cost storage which provides RESTful APIs to store and retrieve objects and their metadata. Key performance indicators (KPIs) of query performance and cost are the number of bytes shipped from Object Storage to Spark and the number of incurred REST requests. We propose Pluggable Spark SQL Filters, which extend the existing Spark SQL partitioning mechanism with an ability to dynamically filter irrelevant objects during query execution. Our approach handles any data format supported by Spark SQL (Parquet, JSON, csv etc.), and unlike pushdown compatible formats such as Parquet which require touching each object to determine its relevance, it avoids accessing irrelevant objects altogether. We developed a pluggable interface for developing and deploying Filters, and implemented GridPocket’s filter which screens objects according to their metadata, for example geo-spatial bounding boxes which describe the area covered by an object’s data points. This leads to drastically lower KPIs since there is no need to ship the entire dataset from Object Storage to Spark if you are only comparing yourself with your neighborhood. We demonstrate GridPocket analytics notebooks, report on our implementation and resulting 10-20x speedups, explain how to implement a Pluggable File Filter, and how we applied this to other use cases.
Optimizing Spark Deployments for Containers: Isolation, Safety, and Performan...Spark Summit
Developers love Linux containers, which neatly package up an application and its dependencies and are easy to create and share. However, this unbeatable developer experience hides some deployment challenges for real applications: how do you wire together pieces of a multi-container application? Where do you store your persistent data if your containers are ephemeral? Do containers really contain and isolate your application, or are they merely hiding potential security vulnerabilities? Are your containers scheduled across your compute resources efficiently, or are they trampling on one another?
Container application platforms like Kubernetes provide the answers to some of these questions. We’ll draw on expertise in Linux security, distributed scheduling, and the Java Virtual Machine to dig deep on the performance and security implications of running in containers. This talk will provide a deep dive into tuning and orchestrating containerized Spark applications. You’ll leave this talk with an understanding of the relevant issues, best practices for containerizing data-processing workloads, and tips for taking advantage of the latest features and fixes in Linux Containers, the JDK, and Kubernetes. You’ll leave inspired and enabled to deploy high-performance Spark applications without giving up the security you need or the developer-friendly workflow you want.
APACHE TOREE: A JUPYTER KERNEL FOR SPARK by Marius van NiekerkSpark Summit
Many data scientists are already making heavy usage of the Jupyter ecosystem for analyzing data using interactive notebooks.
Apache Toree (incubating) is a Jupyter kernel designed to act as a gateway to Spark by enabling users Spark from standard Jupyter notebooks. This allows users to easily integrate Spark into their existing Jupyter deployments, This allows users to easily move between languages and contexts without needing to switch to a different set of tools.
Apache Toree is designed expressly for interactive work. It supports interpreters in Scala, Python, and R.
In this talk, I will cover the design of Toree, how it interacts with the Jupyter ecosystem and various ways in which users can extend the functionality of Apache Toree via a powerful plugin system.
Dask Tutorial at PyConDE / PyData Karlsruhe 2018. These were the introductory slides that mainly contain the link to Matthew Rocklin's Dask workshop at PyData NYC 2018 whereon this workshop was based.
Clipper: A Low-Latency Online Prediction Serving System: Spark Summit East ta...Spark Summit
Machine learning is being deployed in a growing number of applications which demand real-time, accurate, and robust predictions under heavy query load. However, most machine learning frameworks and systems only address model training and not deployment.
In this talk, we present Clipper, a general-purpose low-latency prediction serving system. Interposing between end-user applications and a wide range of machine learning frameworks, Clipper introduces a modular architecture to simplify model deployment across frameworks. Furthermore, by introducing caching, batching, and adaptive model selection techniques, Clipper reduces prediction latency and improves prediction throughput, accuracy, and robustness without modifying the underlying machine learning frameworks. We evaluated Clipper on four common machine learning benchmark datasets and demonstrate its ability to meet the latency, accuracy, and throughput demands of online serving applications. We also compared Clipper to the Tensorflow Serving system and demonstrate comparable prediction throughput and latency on a range of models while enabling new functionality, improved accuracy, and robustness.
How to Choose a Deep Learning FrameworkNavid Kalaei
The trend of neural networks has been attracted a huge community of researchers and practitioners. However, not all of the upfront runners are masters of deep learning and the colorful frameworks could be confusing, especially for the newcomers. In this presentation, I demystified the mystery of the leading frameworks of deep learning and provided a guideline on how to choose the most suitable option.
BigDL: Bringing Ease of Use of Deep Learning for Apache Spark with Jason Dai ...Databricks
BigDL is a distributed deep learning framework for Apache Spark open sourced by Intel. BigDL helps make deep learning more accessible to the Big Data community, by allowing them to continue the use of familiar tools and infrastructure to build deep learning applications. With BigDL, users can write their deep learning applications as standard Spark programs, which can then directly run on top of existing Spark or Hadoop clusters.
In this session, we will introduce BigDL, how our customers use BigDL to build End to End ML/DL applications, platforms on which BigDL is deployed and also provide an update on the latest improvements in BigDL v0.1, and talk about further developments and new upcoming features of BigDL v0.2 release (e.g., support for TensorFlow models, 3D convolutions, etc.).
RISELab:Enabling Intelligent Real-Time DecisionsJen Aman
Spark Summit East Keynote by Ion Stoica
A long-standing grand challenge in computing is to enable machines to act autonomously and intelligently: to rapidly and repeatedly take appropriate actions based on information in the world around them. To address this challenge, at UC Berkeley we are starting a new five year effort that focuses on the development of data-intensive systems that provide Real-Time Intelligence with Secure Execution (RISE). Following in the footsteps of AMPLab, RISELab is an interdisciplinary effort bringing together researchers across AI, robotics, security, and data systems. In this talk I’ll present our research vision and then discuss some of the applications that will be enabled by RISE technologies.
A gentle introduction to Apache Spark from the theorem of Resilient Distributed Datasets to deploying software to the core platform, Spark Streaming, and Spark SQL
Spark-Streaming-as-a-Service with Kafka and YARN: Spark Summit East talk by J...Spark Summit
Since April 2016, Spark-as-a-service has been available to researchers in Sweden from the Swedish ICT SICS Data Center at www.hops.site. Researchers work in an entirely UI-driven environment on a platform built with only open-source software.
Spark applications can be either deployed as jobs (batch or streaming) or written and run directly from Apache Zeppelin. Spark applications are run within a project on a YARN cluster with the novel property that Spark applications are metered and charged to projects. Projects are also securely isolated from each other and include support for project-specific Kafka topics. That is, Kafka topics are protected from access by users that are not members of the project. In this talk we will discuss the challenges in building multi-tenant Spark streaming applications on YARN that are metered and easy-to-debug. We show how we use the ELK stack (Elasticsearch, Logstash, and Kibana) for logging and debugging running Spark streaming applications, how we use Graphana and Graphite for monitoring Spark streaming applications, and how users can debug and optimize terminated Spark Streaming jobs using Dr Elephant. We will also discuss the experiences of our users (over 120 users as of Sept 2016): how they manage their Kafka topics and quotas, patterns for how users share topics between projects, and our novel solutions for helping researchers debug and optimize Spark applications.
To conclude, we will also give an overview on our course ID2223 on Large Scale Learning and Deep Learning, in which 60 students designed and ran SparkML applications on the platform.
Deep Learning to Big Data Analytics on Apache Spark Using BigDL with Xianyan ...Databricks
With the continued success of deep learning techniques, there’s been a rapid growth in applications for perception in many modalities, such as image classification, object detection and speech recognition. In response, Intel’s BigDL is an open source distributed deep learning framework for Apache Spark that includes rich deep learning support and Intel Math Kernel Library acceleration, allowing users to quickly develop deep learning applications with extremely high performance on their existing Hadoop ecosystems.
This sessions will explore several key deep learning applications that Intel successfully built on top of Apache Spark with BigDL. Hear about the technologies they developed and what they learned from building such applications, including: the tool stack in the system and design considerations; an application on image recognition and object detection (faster-rcnn using VGG and PVANET); and an application on speech recognition with deep speech and acoustic feature transformers. He’ll also share other insights and experiences Intel gained while building a unified data analytics platform with Apache Spark MLlib and BigDL.
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...Spark Summit
Here we present a general supervised framework for record deduplication and author-disambiguation via Spark. This work differentiates itself by – Application of Databricks and AWS makes this a scalable implementation. Compute resources are comparably lower than traditional legacy technology using big boxes 24/7. Scalability is crucial as Elsevier’s Scopus data, the biggest scientific abstract repository, covers roughly 250 million authorships from 70 million abstracts covering a few hundred years. – We create a fingerprint for each content by deep learning and/or word2vec algorithms to expedite pairwise similarity calculation. These encoders substantially reduce compute time while maintaining semantic similarity (unlike traditional TFIDF or predefined taxonomies). We will briefly discuss how to optimize word2vec training with high parallelization. Moreover, we show how these encoders can be used to derive a standard representation for all our entities namely such as documents, authors, users, journals, etc. This standard representation can simplify the recommendation problem into a pairwise similarity search and hence it can offer a basic recommender for cross-product applications where we may not have a dedicate recommender engine designed. – Traditional author-disambiguation or record deduplication algorithms are batch-processing with small to no training data. However, we have roughly 25 million authorships that are manually curated or corrected upon user feedback. Hence, it is crucial to maintain historical profiles and hence we have developed a machine learning implementation to deal with data streams and process them in mini batches or one document at a time. We will discuss how to measure the accuracy of such a system, how to tune it and how to process the raw data of pairwise similarity function into final clusters. Lessons learned from this talk can help all sort of companies where they want to integrate their data or deduplicate their user/customer/product databases.
Using Pluggable Apache Spark SQL Filters to Help GridPocket Users Keep Up wit...Spark Summit
Analyzing and comparing your energy consumption with that of other consumers provides healthy peer pressure and useful insight leading to energy conservation and impacting the bottom line. We helped GridPocket (http://www.gridpocket.com/), a smart grid company developing energy management applications for electricity water and gas utilities, implement high scale anonymized energy comparison queries with an order of magnitude lower cost and higher performance than was previously possible. IoT use cases like that of GridPocket are swamping our planet with data, and drive demand for analytics on extremely scalable and low cost storage. Enter Spark SQL over Object Storage: highly scalable and low cost storage which provides RESTful APIs to store and retrieve objects and their metadata. Key performance indicators (KPIs) of query performance and cost are the number of bytes shipped from Object Storage to Spark and the number of incurred REST requests. We propose Pluggable Spark SQL Filters, which extend the existing Spark SQL partitioning mechanism with an ability to dynamically filter irrelevant objects during query execution. Our approach handles any data format supported by Spark SQL (Parquet, JSON, csv etc.), and unlike pushdown compatible formats such as Parquet which require touching each object to determine its relevance, it avoids accessing irrelevant objects altogether. We developed a pluggable interface for developing and deploying Filters, and implemented GridPocket’s filter which screens objects according to their metadata, for example geo-spatial bounding boxes which describe the area covered by an object’s data points. This leads to drastically lower KPIs since there is no need to ship the entire dataset from Object Storage to Spark if you are only comparing yourself with your neighborhood. We demonstrate GridPocket analytics notebooks, report on our implementation and resulting 10-20x speedups, explain how to implement a Pluggable File Filter, and how we applied this to other use cases.
Optimizing Spark Deployments for Containers: Isolation, Safety, and Performan...Spark Summit
Developers love Linux containers, which neatly package up an application and its dependencies and are easy to create and share. However, this unbeatable developer experience hides some deployment challenges for real applications: how do you wire together pieces of a multi-container application? Where do you store your persistent data if your containers are ephemeral? Do containers really contain and isolate your application, or are they merely hiding potential security vulnerabilities? Are your containers scheduled across your compute resources efficiently, or are they trampling on one another?
Container application platforms like Kubernetes provide the answers to some of these questions. We’ll draw on expertise in Linux security, distributed scheduling, and the Java Virtual Machine to dig deep on the performance and security implications of running in containers. This talk will provide a deep dive into tuning and orchestrating containerized Spark applications. You’ll leave this talk with an understanding of the relevant issues, best practices for containerizing data-processing workloads, and tips for taking advantage of the latest features and fixes in Linux Containers, the JDK, and Kubernetes. You’ll leave inspired and enabled to deploy high-performance Spark applications without giving up the security you need or the developer-friendly workflow you want.
APACHE TOREE: A JUPYTER KERNEL FOR SPARK by Marius van NiekerkSpark Summit
Many data scientists are already making heavy usage of the Jupyter ecosystem for analyzing data using interactive notebooks.
Apache Toree (incubating) is a Jupyter kernel designed to act as a gateway to Spark by enabling users Spark from standard Jupyter notebooks. This allows users to easily integrate Spark into their existing Jupyter deployments, This allows users to easily move between languages and contexts without needing to switch to a different set of tools.
Apache Toree is designed expressly for interactive work. It supports interpreters in Scala, Python, and R.
In this talk, I will cover the design of Toree, how it interacts with the Jupyter ecosystem and various ways in which users can extend the functionality of Apache Toree via a powerful plugin system.
Dask Tutorial at PyConDE / PyData Karlsruhe 2018. These were the introductory slides that mainly contain the link to Matthew Rocklin's Dask workshop at PyData NYC 2018 whereon this workshop was based.
Clipper: A Low-Latency Online Prediction Serving System: Spark Summit East ta...Spark Summit
Machine learning is being deployed in a growing number of applications which demand real-time, accurate, and robust predictions under heavy query load. However, most machine learning frameworks and systems only address model training and not deployment.
In this talk, we present Clipper, a general-purpose low-latency prediction serving system. Interposing between end-user applications and a wide range of machine learning frameworks, Clipper introduces a modular architecture to simplify model deployment across frameworks. Furthermore, by introducing caching, batching, and adaptive model selection techniques, Clipper reduces prediction latency and improves prediction throughput, accuracy, and robustness without modifying the underlying machine learning frameworks. We evaluated Clipper on four common machine learning benchmark datasets and demonstrate its ability to meet the latency, accuracy, and throughput demands of online serving applications. We also compared Clipper to the Tensorflow Serving system and demonstrate comparable prediction throughput and latency on a range of models while enabling new functionality, improved accuracy, and robustness.
How to Choose a Deep Learning FrameworkNavid Kalaei
The trend of neural networks has been attracted a huge community of researchers and practitioners. However, not all of the upfront runners are masters of deep learning and the colorful frameworks could be confusing, especially for the newcomers. In this presentation, I demystified the mystery of the leading frameworks of deep learning and provided a guideline on how to choose the most suitable option.
BigDL: Bringing Ease of Use of Deep Learning for Apache Spark with Jason Dai ...Databricks
BigDL is a distributed deep learning framework for Apache Spark open sourced by Intel. BigDL helps make deep learning more accessible to the Big Data community, by allowing them to continue the use of familiar tools and infrastructure to build deep learning applications. With BigDL, users can write their deep learning applications as standard Spark programs, which can then directly run on top of existing Spark or Hadoop clusters.
In this session, we will introduce BigDL, how our customers use BigDL to build End to End ML/DL applications, platforms on which BigDL is deployed and also provide an update on the latest improvements in BigDL v0.1, and talk about further developments and new upcoming features of BigDL v0.2 release (e.g., support for TensorFlow models, 3D convolutions, etc.).
RISELab:Enabling Intelligent Real-Time DecisionsJen Aman
Spark Summit East Keynote by Ion Stoica
A long-standing grand challenge in computing is to enable machines to act autonomously and intelligently: to rapidly and repeatedly take appropriate actions based on information in the world around them. To address this challenge, at UC Berkeley we are starting a new five year effort that focuses on the development of data-intensive systems that provide Real-Time Intelligence with Secure Execution (RISE). Following in the footsteps of AMPLab, RISELab is an interdisciplinary effort bringing together researchers across AI, robotics, security, and data systems. In this talk I’ll present our research vision and then discuss some of the applications that will be enabled by RISE technologies.
A gentle introduction to Apache Spark from the theorem of Resilient Distributed Datasets to deploying software to the core platform, Spark Streaming, and Spark SQL
Spark-Streaming-as-a-Service with Kafka and YARN: Spark Summit East talk by J...Spark Summit
Since April 2016, Spark-as-a-service has been available to researchers in Sweden from the Swedish ICT SICS Data Center at www.hops.site. Researchers work in an entirely UI-driven environment on a platform built with only open-source software.
Spark applications can be either deployed as jobs (batch or streaming) or written and run directly from Apache Zeppelin. Spark applications are run within a project on a YARN cluster with the novel property that Spark applications are metered and charged to projects. Projects are also securely isolated from each other and include support for project-specific Kafka topics. That is, Kafka topics are protected from access by users that are not members of the project. In this talk we will discuss the challenges in building multi-tenant Spark streaming applications on YARN that are metered and easy-to-debug. We show how we use the ELK stack (Elasticsearch, Logstash, and Kibana) for logging and debugging running Spark streaming applications, how we use Graphana and Graphite for monitoring Spark streaming applications, and how users can debug and optimize terminated Spark Streaming jobs using Dr Elephant. We will also discuss the experiences of our users (over 120 users as of Sept 2016): how they manage their Kafka topics and quotas, patterns for how users share topics between projects, and our novel solutions for helping researchers debug and optimize Spark applications.
To conclude, we will also give an overview on our course ID2223 on Large Scale Learning and Deep Learning, in which 60 students designed and ran SparkML applications on the platform.
Deep Learning to Big Data Analytics on Apache Spark Using BigDL with Xianyan ...Databricks
With the continued success of deep learning techniques, there’s been a rapid growth in applications for perception in many modalities, such as image classification, object detection and speech recognition. In response, Intel’s BigDL is an open source distributed deep learning framework for Apache Spark that includes rich deep learning support and Intel Math Kernel Library acceleration, allowing users to quickly develop deep learning applications with extremely high performance on their existing Hadoop ecosystems.
This sessions will explore several key deep learning applications that Intel successfully built on top of Apache Spark with BigDL. Hear about the technologies they developed and what they learned from building such applications, including: the tool stack in the system and design considerations; an application on image recognition and object detection (faster-rcnn using VGG and PVANET); and an application on speech recognition with deep speech and acoustic feature transformers. He’ll also share other insights and experiences Intel gained while building a unified data analytics platform with Apache Spark MLlib and BigDL.
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...Spark Summit
Here we present a general supervised framework for record deduplication and author-disambiguation via Spark. This work differentiates itself by – Application of Databricks and AWS makes this a scalable implementation. Compute resources are comparably lower than traditional legacy technology using big boxes 24/7. Scalability is crucial as Elsevier’s Scopus data, the biggest scientific abstract repository, covers roughly 250 million authorships from 70 million abstracts covering a few hundred years. – We create a fingerprint for each content by deep learning and/or word2vec algorithms to expedite pairwise similarity calculation. These encoders substantially reduce compute time while maintaining semantic similarity (unlike traditional TFIDF or predefined taxonomies). We will briefly discuss how to optimize word2vec training with high parallelization. Moreover, we show how these encoders can be used to derive a standard representation for all our entities namely such as documents, authors, users, journals, etc. This standard representation can simplify the recommendation problem into a pairwise similarity search and hence it can offer a basic recommender for cross-product applications where we may not have a dedicate recommender engine designed. – Traditional author-disambiguation or record deduplication algorithms are batch-processing with small to no training data. However, we have roughly 25 million authorships that are manually curated or corrected upon user feedback. Hence, it is crucial to maintain historical profiles and hence we have developed a machine learning implementation to deal with data streams and process them in mini batches or one document at a time. We will discuss how to measure the accuracy of such a system, how to tune it and how to process the raw data of pairwise similarity function into final clusters. Lessons learned from this talk can help all sort of companies where they want to integrate their data or deduplicate their user/customer/product databases.
Distributed deep rl on spark strata singaporeAdam Gibson
This talk briefly covers deep reinforcemeent learning on spark and the benefits of using large scale commodity compute with gpus for ease of running simulations as well as distributed training for use cases that aren't games such as network intrusion and risk. This talk also briefly mentions rl4j and our work with openai gym.
This talk was on deep learning use cases outside of computer vision. It also covered larger scale patterns of what good deep learning use cases typically look like. We end up on an explanation of anomaly detection and various kinds of anomaly use cases.
Deep learning in production with the bestAdam Gibson
Getting deep learning adopted at your company. The current landscape of academia vs industry. Presentation at AI with the best (online conference):
http://ai.withthebest.com/
Strata Beijing - Deep Learning in Production on SparkAdam Gibson
Recent talk at strata beijing - half english half chinese covering use cases of deep learning, deep learning in production and the different components of deeplearning4j.
Hortonworks - What's Possible with a Modern Data Architecture?Hortonworks
This is Mark Ledbetter's presentation from the September 22, 2014 Hortonworks webinar “What’s Possible with a Modern Data Architecture?” Mark is vice president for industry solutions at Hortonworks. He has more than twenty-five years experience in the software industry with a focus on Retail and supply chain.
Hortonworks and Red Hat Webinar - Part 2Hortonworks
Learn more about creating reference architectures that optimize the delivery the Hortonworks Data Platform. You will hear more about Hive, JBoss Data Virtualization Security, and you will also see in action how to combine sentiment data from Hadoop with data from traditional relational sources.
[Hortonworks] Future Of Data: Madrid - HDF & Data in motionRaúl Marín
First meetup event Future Of Data event where we introduced Hortonworks DataFlow (HDF).
The slides describe what HDF is, and we presented a very simple demo about sentiment analysis of tweets using Apache OpenNLP as the NLP framework to do so.
This webinar series covers Apache Kafka and Apache Storm for streaming data processing. Also, it discusses new streaming innovations for Kafka and Storm included in HDP 2.2
You got your cluster installed and configured. You celebrate, until the party is ruined by your company's Security officer stamping a big "Deny" on your Hadoop cluster. And oops!! You cannot place any data onto the cluster until you can demonstrate it is secure In this session you would learn the tips and tricks to fully secure your cluster for data at rest, data in motion and all the apps including Spark. Your Security officer can then join your Hadoop revelry (unless you don't authorize him to, with your newly acquired admin rights)
Supporting Financial Services with a More Flexible Approach to Big DataHortonworks
Financial services companies can reap tremendous benefits from 'Big Data' and they have moved quickly to deploy it. But these companies also place heavy demands on 'Big Data' infrastructure for flexibility, reliability and performance. In this webinar, Hortonworks joins WANDisco to look at three examples of using 'Big Data' to get a more comprehensive view of customer behavior and activity in the banking and insurance industries. Then we'll pull out the common threads from these examples, and see how a flexible next-generation Hadoop architecture lets you get a step up on improving your business performance. Join us to learn:
How to leverage data from across an entire global enterprise
How to analyze a wide variety of structured and unstructured data to get quick, meaningful answers to critical questions
What industry leaders have put in place
Mr. Slim Baltagi is a Systems Architect at Hortonworks, with over 4 years of Hadoop experience working on 9 Big Data projects: Advanced Customer Analytics, Supply Chain Analytics, Medical Coverage Discovery, Payment Plan Recommender, Research Driven Call List for Sales, Prime Reporting Platform, Customer Hub, Telematics, Historical Data Platform; with Fortune 100 clients and global companies from Financial Services, Insurance, Healthcare and Retail.
Mr. Slim Baltagi has worked in various architecture, design, development and consulting roles at.
Accenture, CME Group, TransUnion, Syntel, Allstate, TransAmerica, Credit Suisse, Chicago Board Options Exchange, Federal Reserve Bank of Chicago, CNA, Sears, USG, ACNielsen, Deutshe Bahn.
Mr. Baltagi has also over 14 years of IT experience with an emphasis on full life cycle development of Enterprise Web applications using Java and Open-Source software. He holds a master’s degree in mathematics and is an ABD in computer science from Université Laval, Québec, Canada.
Languages: Java, Python, JRuby, JEE , PHP, SQL, HTML, XML, XSLT, XQuery, JavaScript, UML, JSON
Databases: Oracle, MS SQL Server, MYSQL, PostreSQL
Software: Eclipse, IBM RAD, JUnit, JMeter, YourKit, PVCS, CVS, UltraEdit, Toad, ClearCase, Maven, iText, Visio, Japser Reports, Alfresco, Yslow, Terracotta, Toad, SoapUI, Dozer, Sonar, Git
Frameworks: Spring, Struts, AppFuse, SiteMesh, Tiles, Hibernate, Axis, Selenium RC, DWR Ajax , Xstream
Distributed Computing/Big Data: Hadoop, MapReduce, HDFS, Hive, Pig, Sqoop, HBase, R, RHadoop, Cloudera CDH4, MapR M7, Hortonworks HDP 2.1
Hortonworks tech workshop in-memory processing with sparkHortonworks
Apache Spark offers unique in-memory capabilities and is well suited to a wide variety of data processing workloads including machine learning and micro-batch processing. With HDP 2.2, Apache Spark is a fully supported component of the Hortonworks Data Platform. In this session we will cover the key fundamentals of Apache Spark and operational best practices for executing Spark jobs along with the rest of Big Data workloads. We will also provide a working example to showcase micro-batch and machine learning processing using Apache Spark.
Webinar: Selecting the Right SQL-on-Hadoop SolutionMapR Technologies
In the crowded SQL-on-Hadoop market, choosing the right solution for your business can be difficult. In this webinar, learn firsthand from Rick van der Lans, independent analyst and managing director of R20/Consultancy, how to sort through this market complexity and what tough questions to ask when evaluating perspective SQL-on-Hadoop solutions.
Discover hdp 2.2: Data storage innovations in Hadoop Distributed Filesystem (...Hortonworks
Hortonworks Data Platform 2.2 include HDFS for data storage . In this 30-minute webinar, we discussed data storage innovations, including Heterogeneous storage, encryption, and operational security enhancements.
These slides to the Discover HDP 2.2 Webinar Series: Data Storage Innovations in HDFS explore Heterogeneous storage, Data Encryption and Operational security.
Apache Ambari is a single framework for IT administrators to provision, manage and monitor a Hadoop cluster. Apache Ambari 1.7.0 is included with Hortonworks Data Platform 2.2.
In this 30-minute webinar, Hortonworks Product Manager Jeff Sposetti and Apache Ambari committer Mahadev Konar discussed new capabilities including:
Improvements to Ambari core - such as support for ResourceManager HA
Extensions to Ambari platform - introducing Ambari Administration and Ambari Views
Enhancements to Ambari Stacks - dynamic configuration recommendations and validations via a "Stack Advisor"
Deploying signature verification with deep learningAdam Gibson
Presentation covered building a signature verification system and deploying it to production. This includes resources usage as well as how the model was picked.
Meetup held in Tokyo with Deep learning Otemachi.
Self driving computers active learning workflows with human interpretable ve...Adam Gibson
Human in the loop learning workflows leveraging deep learning to group and cluster data. Also, techniques for accounting for machine learning failures.
Anomaly Detection and Automatic Labeling with Deep LearningAdam Gibson
Adam Gibson demonstrates how to use variational autoencoders to automatically label time series location data. You'll explore the challenge of imbalanced classes and anomaly detection, learn how to leverage deep learning for automatically labeling (and the pitfalls of this), and discover how you can deploy these techniques in your organization.
Recent presentation on deeplearning4j's new features as well as some underused features of the AI framework like arbiter,datavec's transform process and libnd4j.
Gave a talk at:
www.meetup.com/SF-Bayarea-Machine-Learning/events/221739934/
Covers basic architecture of a scientific lib and my take on it with nd4j.
These slides accompanied a demo of Deeplearning4j at the SF Data Mining Meetup hosted by Trulia.
http://www.meetup.com/Data-Mining/events/212445872/
Deep-learning is useful in detecting identifying similarities to augment search and text analytics; predicting customer lifetime value and churn; and recognizing faces and voices.
Deeplearning4j is an infinitely scalable deep-learning architecture suitable for Hadoop and other big-data structures. It includes a distributed deep-learning framework and a normal deep-learning framework; i.e. it runs on a single thread as well. Training takes place in the cluster, which means it can process massive amounts of data. Nets are trained in parallel via iterative reduce, and they are equally compatible with Java, Scala and Clojure. The distributed deep-learning framework is made for data input and neural net training at scale, and its output should be highly accurate predictive models.
The framework's neural nets include restricted Boltzmann machines, deep-belief networks, deep autoencoders, convolutional nets and recursive neural tensor networks.
Finally, Deeplearning4j integrates with GPUs. A stable version was released in October.
Show drafts
volume_up
Empowering the Data Analytics Ecosystem: A Laser Focus on Value
The data analytics ecosystem thrives when every component functions at its peak, unlocking the true potential of data. Here's a laser focus on key areas for an empowered ecosystem:
1. Democratize Access, Not Data:
Granular Access Controls: Provide users with self-service tools tailored to their specific needs, preventing data overload and misuse.
Data Catalogs: Implement robust data catalogs for easy discovery and understanding of available data sources.
2. Foster Collaboration with Clear Roles:
Data Mesh Architecture: Break down data silos by creating a distributed data ownership model with clear ownership and responsibilities.
Collaborative Workspaces: Utilize interactive platforms where data scientists, analysts, and domain experts can work seamlessly together.
3. Leverage Advanced Analytics Strategically:
AI-powered Automation: Automate repetitive tasks like data cleaning and feature engineering, freeing up data talent for higher-level analysis.
Right-Tool Selection: Strategically choose the most effective advanced analytics techniques (e.g., AI, ML) based on specific business problems.
4. Prioritize Data Quality with Automation:
Automated Data Validation: Implement automated data quality checks to identify and rectify errors at the source, minimizing downstream issues.
Data Lineage Tracking: Track the flow of data throughout the ecosystem, ensuring transparency and facilitating root cause analysis for errors.
5. Cultivate a Data-Driven Mindset:
Metrics-Driven Performance Management: Align KPIs and performance metrics with data-driven insights to ensure actionable decision making.
Data Storytelling Workshops: Equip stakeholders with the skills to translate complex data findings into compelling narratives that drive action.
Benefits of a Precise Ecosystem:
Sharpened Focus: Precise access and clear roles ensure everyone works with the most relevant data, maximizing efficiency.
Actionable Insights: Strategic analytics and automated quality checks lead to more reliable and actionable data insights.
Continuous Improvement: Data-driven performance management fosters a culture of learning and continuous improvement.
Sustainable Growth: Empowered by data, organizations can make informed decisions to drive sustainable growth and innovation.
By focusing on these precise actions, organizations can create an empowered data analytics ecosystem that delivers real value by driving data-driven decisions and maximizing the return on their data investment.
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...Subhajit Sahu
Abstract — Levelwise PageRank is an alternative method of PageRank computation which decomposes the input graph into a directed acyclic block-graph of strongly connected components, and processes them in topological order, one level at a time. This enables calculation for ranks in a distributed fashion without per-iteration communication, unlike the standard method where all vertices are processed in each iteration. It however comes with a precondition of the absence of dead ends in the input graph. Here, the native non-distributed performance of Levelwise PageRank was compared against Monolithic PageRank on a CPU as well as a GPU. To ensure a fair comparison, Monolithic PageRank was also performed on a graph where vertices were split by components. Results indicate that Levelwise PageRank is about as fast as Monolithic PageRank on the CPU, but quite a bit slower on the GPU. Slowdown on the GPU is likely caused by a large submission of small workloads, and expected to be non-issue when the computation is performed on massive graphs.
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...pchutichetpong
M Capital Group (“MCG”) expects to see demand and the changing evolution of supply, facilitated through institutional investment rotation out of offices and into work from home (“WFH”), while the ever-expanding need for data storage as global internet usage expands, with experts predicting 5.3 billion users by 2023. These market factors will be underpinned by technological changes, such as progressing cloud services and edge sites, allowing the industry to see strong expected annual growth of 13% over the next 4 years.
Whilst competitive headwinds remain, represented through the recent second bankruptcy filing of Sungard, which blames “COVID-19 and other macroeconomic trends including delayed customer spending decisions, insourcing and reductions in IT spending, energy inflation and reduction in demand for certain services”, the industry has seen key adjustments, where MCG believes that engineering cost management and technological innovation will be paramount to success.
MCG reports that the more favorable market conditions expected over the next few years, helped by the winding down of pandemic restrictions and a hybrid working environment will be driving market momentum forward. The continuous injection of capital by alternative investment firms, as well as the growing infrastructural investment from cloud service providers and social media companies, whose revenues are expected to grow over 3.6x larger by value in 2026, will likely help propel center provision and innovation. These factors paint a promising picture for the industry players that offset rising input costs and adapt to new technologies.
According to M Capital Group: “Specifically, the long-term cost-saving opportunities available from the rise of remote managing will likely aid value growth for the industry. Through margin optimization and further availability of capital for reinvestment, strong players will maintain their competitive foothold, while weaker players exit the market to balance supply and demand.”
Opendatabay - Open Data Marketplace.pptxOpendatabay
Opendatabay.com unlocks the power of data for everyone. Open Data Marketplace fosters a collaborative hub for data enthusiasts to explore, share, and contribute to a vast collection of datasets.
First ever open hub for data enthusiasts to collaborate and innovate. A platform to explore, share, and contribute to a vast collection of datasets. Through robust quality control and innovative technologies like blockchain verification, opendatabay ensures the authenticity and reliability of datasets, empowering users to make data-driven decisions with confidence. Leverage cutting-edge AI technologies to enhance the data exploration, analysis, and discovery experience.
From intelligent search and recommendations to automated data productisation and quotation, Opendatabay AI-driven features streamline the data workflow. Finding the data you need shouldn't be a complex. Opendatabay simplifies the data acquisition process with an intuitive interface and robust search tools. Effortlessly explore, discover, and access the data you need, allowing you to focus on extracting valuable insights. Opendatabay breaks new ground with a dedicated, AI-generated, synthetic datasets.
Leverage these privacy-preserving datasets for training and testing AI models without compromising sensitive information. Opendatabay prioritizes transparency by providing detailed metadata, provenance information, and usage guidelines for each dataset, ensuring users have a comprehensive understanding of the data they're working with. By leveraging a powerful combination of distributed ledger technology and rigorous third-party audits Opendatabay ensures the authenticity and reliability of every dataset. Security is at the core of Opendatabay. Marketplace implements stringent security measures, including encryption, access controls, and regular vulnerability assessments, to safeguard your data and protect your privacy.
8. How do we realize MDA in a Hadoop Centric World?
HDF
Hadoop
HDFS
HBase Hive SOLR
YARN
Storm
Service
Management /
Workflow
SIEM
Spark
Raw Network Stream
Network Metadata Stream
Data Stores
Syslog
Raw Application Logs
Other Streaming Telemetry
TALK TRACK
I’m about to go over the products, consulting and training that Hortonworks offers, and I want you to keep this image in mind.
Remember:
The Internet of Anything is doubling the amount of data in the world every 2 years.
Connected Data Platforms deliver an open-architected solution to manage data, both in motion and at rest, empowering your organization to gain Actionable Intelligence delivered to your end users through Modern Data Apps.
Hortonworks DataFlow (aka HDF) manages your data in motion—bringing it to where you need it for real-time analysis to capture perishable insights or into storage for historical analysis.
Hortonworks Data Platform (aka HDP) stores the data at rest and provides historical insights through deep, detailed analysis of everything that’s already happened.
Those historical insights from HDP help optimize your data ingest with HDF, which in turn optimizes your data at rest.
This is how HDF, HDP, and Modern Data Applications deliver actionable intelligence to your end users.
And Actionable Intelligence is the beating heart animating the Future of Data.
[NEXT SLIDE]
CapOne – Ingesting from everywhere
Email, Syslog, Applog, Netflow…
Moving to “Cloud Only model”….even looking to use “docker Containers” in Amazon…
The team puts together a detailed architecture of the proposed solution using HDP and HDF. The architecture considers sources data from the numerous sources including Server Logs, Application Logs, XML and Senso data. This data is easily accepted into the flexible schema of HDP using HDF and Sqoop. The data is processed using Pig and analyzed using Spark. Then the data is made available in a real-time dashboard as well as to visualization and reporting tools.
[NEXT SLIDE]