More Data, More Problems: Scaling Kafka-Mirroring Pipelines at LinkedIn confluent
(Celia Kung, LinkedIn) Kafka Summit SF 2018
For several years, LinkedIn has been using Kafka MirrorMaker as the mirroring solution for copying data between Kafka clusters across data centers. However, as LinkedIn data continued to grow, mirroring trillions of Kafka messages per day across data centers uncovered the scale limitations and operability challenges of Kafka MirrorMaker. To address such issues, we have developed a new mirroring solution, built on top our stream ingestion service, Brooklin. Brooklin MirrorMaker aims to provide improved performance and stability, while facilitating better management through finer control of data pipelines. Through flushless Kafka produce, dynamic management of data pipelines, per-partition error handling and flow control, we are able to increase throughput, better withstand consume and produce failures and reduce overall operating costs. As a result, we have eliminated the major pain points of Kafka MirrorMaker. In this talk, we will dive deeper into the challenges LinkedIn has faced with Kafka MirrorMaker, how we tackled them with Brooklin MirrorMaker and our plans for iterating further on this new mirroring solution.
Lessons Learned Migrating from IBM BigInsights to Hortonworks Data PlatformDataWorks Summit
Overview of USAAs decision drivers and journey to migrate from our IBM BigInsights Hadoop platform to Hortonworks Data Platform. Many obstacles challenged our ability to respond to demands of our business data needs and analytic capabilities. In addition to migrating 1.5 PB (500 TB useable) of data to our new HDP environment, we were introducing a new security model with Kerberos/AD integration, data governance, as well as many new HDP services that were unavailable within our BigInsights platform. We'll discuss overall scope of work for this year long journey and our approach in bringing enterprise adoption of a new Hadoop platform. We still have many efforts under way to further enhance our data delivery patterns, information governance process and procedures, and optimized consumption within our HDP platform, but we are now better positioned to provide a flexible, secure, and managed Hadoop platform with focused innovation to meet USAAs strategic initiatives.
Speaker
Lisa Coleman, USAA, Technical Architect
Robert Tucker, USAA, Software Developer & Integrator Lead
Flink SQL & TableAPI in Large Scale Production at AlibabaDataWorks Summit
Search and recommendation system for Alibaba’s e-commerce platform use batch and streaming processing heavily. Flink SQL and Table API (which is a SQL-like DSL) provide simple, flexible, and powerful language to express the data processing logic. More importantly, it opens the door to unify the semantics of batch and streaming jobs.
Blink is a project at Alibaba which improves Apache Flink to make it ready for large scale production use. To support our products, we made lots of improvements to Flink SQL & TableAPI in Alibaba's Blink project. We added the support for User-Defined Table function (UDTF), User-Defined Aggregates (UDAGG), Window Aggregate, and retraction, etc. We are actively working with the Flink community to contribute these improvements back. In this talk, we will present the rationale, semantics, design and implementation of these improvements. We will also share the experience of running large scale Flink SQL and TableAPI jobs at Alibaba.
More Data, More Problems: Scaling Kafka-Mirroring Pipelines at LinkedIn confluent
(Celia Kung, LinkedIn) Kafka Summit SF 2018
For several years, LinkedIn has been using Kafka MirrorMaker as the mirroring solution for copying data between Kafka clusters across data centers. However, as LinkedIn data continued to grow, mirroring trillions of Kafka messages per day across data centers uncovered the scale limitations and operability challenges of Kafka MirrorMaker. To address such issues, we have developed a new mirroring solution, built on top our stream ingestion service, Brooklin. Brooklin MirrorMaker aims to provide improved performance and stability, while facilitating better management through finer control of data pipelines. Through flushless Kafka produce, dynamic management of data pipelines, per-partition error handling and flow control, we are able to increase throughput, better withstand consume and produce failures and reduce overall operating costs. As a result, we have eliminated the major pain points of Kafka MirrorMaker. In this talk, we will dive deeper into the challenges LinkedIn has faced with Kafka MirrorMaker, how we tackled them with Brooklin MirrorMaker and our plans for iterating further on this new mirroring solution.
Lessons Learned Migrating from IBM BigInsights to Hortonworks Data PlatformDataWorks Summit
Overview of USAAs decision drivers and journey to migrate from our IBM BigInsights Hadoop platform to Hortonworks Data Platform. Many obstacles challenged our ability to respond to demands of our business data needs and analytic capabilities. In addition to migrating 1.5 PB (500 TB useable) of data to our new HDP environment, we were introducing a new security model with Kerberos/AD integration, data governance, as well as many new HDP services that were unavailable within our BigInsights platform. We'll discuss overall scope of work for this year long journey and our approach in bringing enterprise adoption of a new Hadoop platform. We still have many efforts under way to further enhance our data delivery patterns, information governance process and procedures, and optimized consumption within our HDP platform, but we are now better positioned to provide a flexible, secure, and managed Hadoop platform with focused innovation to meet USAAs strategic initiatives.
Speaker
Lisa Coleman, USAA, Technical Architect
Robert Tucker, USAA, Software Developer & Integrator Lead
Flink SQL & TableAPI in Large Scale Production at AlibabaDataWorks Summit
Search and recommendation system for Alibaba’s e-commerce platform use batch and streaming processing heavily. Flink SQL and Table API (which is a SQL-like DSL) provide simple, flexible, and powerful language to express the data processing logic. More importantly, it opens the door to unify the semantics of batch and streaming jobs.
Blink is a project at Alibaba which improves Apache Flink to make it ready for large scale production use. To support our products, we made lots of improvements to Flink SQL & TableAPI in Alibaba's Blink project. We added the support for User-Defined Table function (UDTF), User-Defined Aggregates (UDAGG), Window Aggregate, and retraction, etc. We are actively working with the Flink community to contribute these improvements back. In this talk, we will present the rationale, semantics, design and implementation of these improvements. We will also share the experience of running large scale Flink SQL and TableAPI jobs at Alibaba.
Modern ETL Pipelines with Change Data CaptureDatabricks
In this talk we’ll present how at GetYourGuide we’ve built from scratch a completely new ETL pipeline using Debezium, Kafka, Spark and Airflow, which can automatically handle schema changes. Our starting point was an error prone legacy system that ran daily, and was vulnerable to breaking schema changes, which caused many sleepless on-call nights. As most companies, we also have traditional SQL databases that we need to connect to in order to extract relevant data.
This is done usually through either full or partial copies of the data with tools such as sqoop. However another approach that has become quite popular lately is to use Debezium as the Change Data Capture layer which reads databases binlogs, and stream these changes directly to Kafka. As having data once a day is not enough anymore for our bussiness, and we wanted our pipelines to be resilent to upstream schema changes, we’ve decided to rebuild our ETL using Debezium.
We’ll walk the audience through the steps we followed to architect and develop such solution using Databricks to reduce operation time. By building this new pipeline we are now able to refresh our data lake multiple times a day, giving our users fresh data, and protecting our nights of sleep.
Enabling Insight to Support World-Class Supercomputing (Stefan Ceballos, Oak ...confluent
The Oak Ridge Leadership Facility (OLCF) in the National Center for Computational Sciences (NCCS) division at Oak Ridge National Laboratory (ORNL) houses world-class high-performance computing (HPC) resources and has a history of operating top-ranked supercomputers on the TOP500 list, including the world's current fastest, Summit, an IBM AC922 machine with a peak of 200 petaFLOPS. With the exascale era rapidly approaching, the need for a robust and scalable big data platform for operations data is more important than ever. In the past when a new HPC resource was added to the facility, pipelines from data sources spanned multiple data sinks which oftentimes resulted in data silos, slow operational data onboarding, and non-scalable data pipelines for batch processing. Using Apache Kafka as the message bus of the division's new big data platform has allowed for easier decoupling of scalable data pipelines, faster data onboarding, and stream processing with the goal to continuously improve insight into the HPC resources and their supporting systems. This talk will focus on the NCCS division's transition to Apache Kafka over the past few years to enhance the OLCF's current capabilities and prepare for Frontier, OLCF's future exascale system; including the development and deployment of a full big data platform in a Kubernetes environment from both a technical and cultural shift perspective. This talk will also cover the mission of the OLCF, the operational data insights related to high-performance computing that the organization strives for, and several use-cases that exist in production today.
Presto, an open source distributed SQL engine, is widely recognized for its low-latency queries, high concurrency, and native ability to query multiple data sources. Proven at scale in a variety of use cases at Facebook, Airbnb, Netflix, Uber, Twitter, Bloomberg, and FINRA, Presto experienced an unprecedented growth in popularity in both on-premises and cloud deployments in the last few years.
Inspired by the increasingly complex SQL queries run by the Presto user community, engineers at Facebook and Starburst have recently focused on cost-based query optimization. In this talk we will present the initial design and implementation of the CBO, support for connector-provided statistics, estimating selectivity, and choosing efficient query plans. Then, our detailed experimental evaluation will illustrate the performance gains for several classes of queries achieved thanks to the optimizer. Finally, we will discuss our future work enhancing the initial CBO and present the general Presto roadmap for 2018 and beyond.
Speakers
Kamil Bajda-Pawlikowski, Starburst Data, CTO & Co-Founder
Martin Traverso
Stream data processing is increasingly required to support business needs for faster actionable insight with growing volume of information from more sources. Apache Apex is a true stream processing framework for low-latency, high-throughput and reliable processing of complex analytics pipelines on clusters. Apex is designed for quick time-to-production, and is used in production by large companies for real-time and batch processing at scale.
This session will use an Apex production use case to walk through the incremental transition from a batch pipeline with hours of latency to an end-to-end streaming architecture with billions of events per day which are processed to deliver real-time analytical reports. The example is representative for many similar extract-transform-load (ETL) use cases with other data sets that can use a common library of building blocks. The transform (or analytics) piece of such pipelines varies in complexity and often involves business logic specific, custom components.
Topics include:
* Pipeline functionality from event source through queryable state for real-time insights.
* API for application development and development process.
* Library of building blocks including connectors for sources and sinks such as Kafka, JMS, Cassandra, HBase, JDBC and how they enable end-to-end exactly-once results.
* Stateful processing with event time windowing.
* Fault tolerance with exactly-once result semantics, checkpointing, incremental recovery
* Scalability and low-latency, high-throughput processing with advanced engine features for auto-scaling, dynamic changes, compute locality.
* Who is using Apex in production, and roadmap.
Following the session attendees will have a high level understanding of Apex and how it can be applied to use cases at their own organizations.
LinkedIn's Logical Data Access Layer for Hadoop -- Strata London 2016Carl Steinbach
An overview of Dali, LinkedIn's logical data access layer for Hadoop. Dali provides cluster and version-independent access to HDFS filesystems, a dataset API that supports virtualized datasets and dataset versioning, and explicit contract management governing the evolution of datasets.
Apache Pinot Case Study: Building Distributed Analytics Systems Using Apache ...HostedbyConfluent
We built Apache Pinot - a real-time distributed OLAP datastore - for low-latency analytics at scale. This is heavily used at companies such as LinkedIn, Uber, Slack, where Kafka serves as the backbone for capturing vast amounts of data. Pinot ingests millions of events per sec from Kafka, builds indexes in real-time and serves 100K+ queries per second while ensuring latency SLA of millisecond to sub second.
In the first implementation, we used the Consumer Group feature to manage the offsets and checkpoints across multiple Kafka Consumers. However, to achieve fault tolerance and scalability, we had to run multiple consumer groups for the same topic. This was our initial strategy to maintain the SLA at high query workload. But this model posed other challenges - since Kafka maintains offset per consumer group, achieving data consistency across multiple consumer groups was not possible. Also, a failure of a single node in a consumer group meant the entire consumer group was unavailable for query processing. Restarting the failed node needed lot of manual operations to ensure data is consumed exactly once. This resulted in management overhead and inefficient hardware utilization.
While taking inspiration from the Kafka consumer group implementation, we redesigned the real-time consumption in Pinot to maintain consistent offset across multiple consumer groups. This allowed us to guarantee consistent data across all replicas. This enabled us to copy data from another consumer group during node addition, node failure or increasing the replication group.
In this talk, we will deep dive into the various challenges faced and considerations that went into this design, and learn what makes Pinot resilient to failures both in Kafka Brokers and Pinot Components. We will introduce the new concept of ""lockstep"" sequencing where multiple consumer groups can synchronize checkpoints periodically and maintain consistency. We'll describe how we achieve this while maintaining strict freshness SLAs, and also withstanding high throughput and ingestion.
HBaseCon 2015: Industrial Internet Case Study using HBase and TSDBHBaseCon
This case study involves analysis of high-volume, continuous time-series aviation data from jet engines that consist of temperature, pressure, vibration and related parameters from the on-board sensors, joined with well-characterized slowly changing engine asset configuration data and other enterprise data for continuous engine diagnostics and analytics. This data is ingested via distributed fabric comprising transient containers, message queues and a columnar, compressed storage leveraging OpenTSDB over Apache HBase.
Streaming Data Lakes using Kafka Connect + Apache Hudi | Vinoth Chandar, Apac...HostedbyConfluent
Apache Hudi is a data lake platform, that provides streaming primitives (upserts/deletes/change streams) on top of data lake storage. Hudi powers very large data lakes at Uber, Robinhood and other companies, while being pre-installed on four major cloud platforms.
Hudi supports exactly-once, near real-time data ingestion from Apache Kafka to cloud storage, which is typically used in-place of a S3/HDFS sink connector to gain transactions and mutability. While this approach is scalable and battle-tested, it can only ingest data in mini batches, leading to lower data freshness. In this talk, we introduce a Kafka Connect Sink Connector for Apache Hudi, which writes data straight into Hudi's log format, making the data immediately queryable, while Hudi's table services like indexing, compaction, clustering work behind the scenes, to further re-organize for better query performance.
Tapad's data pipeline is an elastic combination of technologies (Kafka, Hadoop, Avro, Scalding) that forms a reliable system for analytics, realtime and batch graph-building, and logging. In this talk, I will speak about the creation and evolution of the pipeline, and a concrete example – a day in the life of an event tracking pixel. We'll also talk about common challenges that we've overcome such as integrating different pieces of the system, schema evolution, queuing, and data retention policies.
Streaming all over the world Real life use cases with Kafka Streamsconfluent
Streaming all over the world Real life use cases with Kafka Streams, Dr. Benedikt Linse, Senior Solutions Architect, Confluent
https://www.meetup.com/Apache-Kafka-Germany-Munich/events/281819704/
Streaming data in the cloud with Confluent and MongoDB Atlas | Robert Waters,...HostedbyConfluent
Are you looking for a cloud-based architecture that includes the best of breed streaming and database technologies? In this session you will learn how to setup and configure the Confluent Cloud with MongoDB Atlas. We'll start the journey learning about the basic connectivity between the two cloud services and end with a brief discovery of what you can do with data once it is in MongoDB Atlas. By the end of this session you will know how to securely setup and configure the MongoDB Atlas connectors in the Confluent Cloud in both a source and sink configuration.
Symantec: Cassandra Data Modelling techniques in actionDataStax Academy
Our product presents an aggregated view of metadata collected for billions of objects (files, emails, sharepoint objects etc.). We used Cassandra to store those billions of objects along with aggregated view of that metadata. Customers can analyse the corpus of data in real time by searching in completely flexible way i.e. be able to get summary aggregates for many billions of objects, and then be able to further drill down to items by filtering using various facets of the metadata. We achieve this using a combination of Cassandra and ElasticSearch. This presentation will talk about various data modelling techniques we use to aggregate and then further summarise all that metadata and be able to search the summary in real t
Embracing Database Diversity with Kafka and DebeziumFrank Lyaruu
There was a time not long ago when we used relational databases for everything. Even if the data wasn’t particularly relational, we shoehorned it into relational tables, often because that was the only database we had. Thank god these dark times are over and now we have many different kinds of NoSQL databases: Document, realtime, graph, column, but that does not solve the problem that the same data might be a graph from one perspective, but a collection of documents from another.
It would be really nice if we can access that same data in many different ways, depending on the context of what we want to achieve in our current task.
As software architects this is not easy to solve but definitely possible: We can design an architecture using Event Sourcing: Capture the data with Debezium, post it to a Kafka queue, use Kafka Streams to model the data the way we like, and store the data in various different data sources, so we can synchronize data between data sources.
Bringing Streaming Data To The Masses: Lowering The “Cost Of Admission” For Y...confluent
(Bob Lehmann, Bayer) Kafka Summit SF 2018
You’ve built your streaming data platform. The early adopters are “all in” and have developed producers, consumers and stream processing apps for a number of use cases. A large percentage of the enterprise, however, has expressed interest but hasn’t made the leap. Why?
In 2014, Bayer Crop Science (formerly Monsanto) adopted a cloud first strategy and started a multi-year transition to the cloud. A Kafka-based cross-datacenter DataHub was created to facilitate this migration and to drive the shift to real-time stream processing. The DataHub has seen strong enterprise adoption and supports a myriad of use cases. Data is ingested from a wide variety of sources and the data can move effortlessly between an on premise datacenter, AWS and Google Cloud. The DataHub has evolved continuously over time to meet the current and anticipated needs of our internal customers. The “cost of admission” for the platform has been lowered dramatically over time via our DataHub Portal and technologies such as Kafka Connect, Kubernetes and Presto. Most operations are now self-service, onboarding of new data sources is relatively painless and stream processing via KSQL and other technologies is being incorporated into the core DataHub platform.
In this talk, Bob Lehmann will describe the origins and evolution of the Enterprise DataHub with an emphasis on steps that were taken to drive user adoption. Bob will also talk about integrations between the DataHub and other key data platforms at Bayer, lessons learned and the future direction for streaming data and stream processing at Bayer.
High Speed Continuous & Reliable Data Ingest into HadoopDataWorks Summit
This talk will explore the area of real-time data ingest into Hadoop and present the architectural trade-offs as well as demonstrate alternative implementations that strike the appropriate balance across the following common challenges: * Decentralized writes (multiple data centers and collectors) * Continuous Availability, High Reliability * No loss of data * Elasticity of introducing more writers * Bursts in Speed per syslog emitter * Continuous, real-time collection * Flexible Write Targets (local FS, HDFS etc.)
Modern ETL Pipelines with Change Data CaptureDatabricks
In this talk we’ll present how at GetYourGuide we’ve built from scratch a completely new ETL pipeline using Debezium, Kafka, Spark and Airflow, which can automatically handle schema changes. Our starting point was an error prone legacy system that ran daily, and was vulnerable to breaking schema changes, which caused many sleepless on-call nights. As most companies, we also have traditional SQL databases that we need to connect to in order to extract relevant data.
This is done usually through either full or partial copies of the data with tools such as sqoop. However another approach that has become quite popular lately is to use Debezium as the Change Data Capture layer which reads databases binlogs, and stream these changes directly to Kafka. As having data once a day is not enough anymore for our bussiness, and we wanted our pipelines to be resilent to upstream schema changes, we’ve decided to rebuild our ETL using Debezium.
We’ll walk the audience through the steps we followed to architect and develop such solution using Databricks to reduce operation time. By building this new pipeline we are now able to refresh our data lake multiple times a day, giving our users fresh data, and protecting our nights of sleep.
Enabling Insight to Support World-Class Supercomputing (Stefan Ceballos, Oak ...confluent
The Oak Ridge Leadership Facility (OLCF) in the National Center for Computational Sciences (NCCS) division at Oak Ridge National Laboratory (ORNL) houses world-class high-performance computing (HPC) resources and has a history of operating top-ranked supercomputers on the TOP500 list, including the world's current fastest, Summit, an IBM AC922 machine with a peak of 200 petaFLOPS. With the exascale era rapidly approaching, the need for a robust and scalable big data platform for operations data is more important than ever. In the past when a new HPC resource was added to the facility, pipelines from data sources spanned multiple data sinks which oftentimes resulted in data silos, slow operational data onboarding, and non-scalable data pipelines for batch processing. Using Apache Kafka as the message bus of the division's new big data platform has allowed for easier decoupling of scalable data pipelines, faster data onboarding, and stream processing with the goal to continuously improve insight into the HPC resources and their supporting systems. This talk will focus on the NCCS division's transition to Apache Kafka over the past few years to enhance the OLCF's current capabilities and prepare for Frontier, OLCF's future exascale system; including the development and deployment of a full big data platform in a Kubernetes environment from both a technical and cultural shift perspective. This talk will also cover the mission of the OLCF, the operational data insights related to high-performance computing that the organization strives for, and several use-cases that exist in production today.
Presto, an open source distributed SQL engine, is widely recognized for its low-latency queries, high concurrency, and native ability to query multiple data sources. Proven at scale in a variety of use cases at Facebook, Airbnb, Netflix, Uber, Twitter, Bloomberg, and FINRA, Presto experienced an unprecedented growth in popularity in both on-premises and cloud deployments in the last few years.
Inspired by the increasingly complex SQL queries run by the Presto user community, engineers at Facebook and Starburst have recently focused on cost-based query optimization. In this talk we will present the initial design and implementation of the CBO, support for connector-provided statistics, estimating selectivity, and choosing efficient query plans. Then, our detailed experimental evaluation will illustrate the performance gains for several classes of queries achieved thanks to the optimizer. Finally, we will discuss our future work enhancing the initial CBO and present the general Presto roadmap for 2018 and beyond.
Speakers
Kamil Bajda-Pawlikowski, Starburst Data, CTO & Co-Founder
Martin Traverso
Stream data processing is increasingly required to support business needs for faster actionable insight with growing volume of information from more sources. Apache Apex is a true stream processing framework for low-latency, high-throughput and reliable processing of complex analytics pipelines on clusters. Apex is designed for quick time-to-production, and is used in production by large companies for real-time and batch processing at scale.
This session will use an Apex production use case to walk through the incremental transition from a batch pipeline with hours of latency to an end-to-end streaming architecture with billions of events per day which are processed to deliver real-time analytical reports. The example is representative for many similar extract-transform-load (ETL) use cases with other data sets that can use a common library of building blocks. The transform (or analytics) piece of such pipelines varies in complexity and often involves business logic specific, custom components.
Topics include:
* Pipeline functionality from event source through queryable state for real-time insights.
* API for application development and development process.
* Library of building blocks including connectors for sources and sinks such as Kafka, JMS, Cassandra, HBase, JDBC and how they enable end-to-end exactly-once results.
* Stateful processing with event time windowing.
* Fault tolerance with exactly-once result semantics, checkpointing, incremental recovery
* Scalability and low-latency, high-throughput processing with advanced engine features for auto-scaling, dynamic changes, compute locality.
* Who is using Apex in production, and roadmap.
Following the session attendees will have a high level understanding of Apex and how it can be applied to use cases at their own organizations.
LinkedIn's Logical Data Access Layer for Hadoop -- Strata London 2016Carl Steinbach
An overview of Dali, LinkedIn's logical data access layer for Hadoop. Dali provides cluster and version-independent access to HDFS filesystems, a dataset API that supports virtualized datasets and dataset versioning, and explicit contract management governing the evolution of datasets.
Apache Pinot Case Study: Building Distributed Analytics Systems Using Apache ...HostedbyConfluent
We built Apache Pinot - a real-time distributed OLAP datastore - for low-latency analytics at scale. This is heavily used at companies such as LinkedIn, Uber, Slack, where Kafka serves as the backbone for capturing vast amounts of data. Pinot ingests millions of events per sec from Kafka, builds indexes in real-time and serves 100K+ queries per second while ensuring latency SLA of millisecond to sub second.
In the first implementation, we used the Consumer Group feature to manage the offsets and checkpoints across multiple Kafka Consumers. However, to achieve fault tolerance and scalability, we had to run multiple consumer groups for the same topic. This was our initial strategy to maintain the SLA at high query workload. But this model posed other challenges - since Kafka maintains offset per consumer group, achieving data consistency across multiple consumer groups was not possible. Also, a failure of a single node in a consumer group meant the entire consumer group was unavailable for query processing. Restarting the failed node needed lot of manual operations to ensure data is consumed exactly once. This resulted in management overhead and inefficient hardware utilization.
While taking inspiration from the Kafka consumer group implementation, we redesigned the real-time consumption in Pinot to maintain consistent offset across multiple consumer groups. This allowed us to guarantee consistent data across all replicas. This enabled us to copy data from another consumer group during node addition, node failure or increasing the replication group.
In this talk, we will deep dive into the various challenges faced and considerations that went into this design, and learn what makes Pinot resilient to failures both in Kafka Brokers and Pinot Components. We will introduce the new concept of ""lockstep"" sequencing where multiple consumer groups can synchronize checkpoints periodically and maintain consistency. We'll describe how we achieve this while maintaining strict freshness SLAs, and also withstanding high throughput and ingestion.
HBaseCon 2015: Industrial Internet Case Study using HBase and TSDBHBaseCon
This case study involves analysis of high-volume, continuous time-series aviation data from jet engines that consist of temperature, pressure, vibration and related parameters from the on-board sensors, joined with well-characterized slowly changing engine asset configuration data and other enterprise data for continuous engine diagnostics and analytics. This data is ingested via distributed fabric comprising transient containers, message queues and a columnar, compressed storage leveraging OpenTSDB over Apache HBase.
Streaming Data Lakes using Kafka Connect + Apache Hudi | Vinoth Chandar, Apac...HostedbyConfluent
Apache Hudi is a data lake platform, that provides streaming primitives (upserts/deletes/change streams) on top of data lake storage. Hudi powers very large data lakes at Uber, Robinhood and other companies, while being pre-installed on four major cloud platforms.
Hudi supports exactly-once, near real-time data ingestion from Apache Kafka to cloud storage, which is typically used in-place of a S3/HDFS sink connector to gain transactions and mutability. While this approach is scalable and battle-tested, it can only ingest data in mini batches, leading to lower data freshness. In this talk, we introduce a Kafka Connect Sink Connector for Apache Hudi, which writes data straight into Hudi's log format, making the data immediately queryable, while Hudi's table services like indexing, compaction, clustering work behind the scenes, to further re-organize for better query performance.
Tapad's data pipeline is an elastic combination of technologies (Kafka, Hadoop, Avro, Scalding) that forms a reliable system for analytics, realtime and batch graph-building, and logging. In this talk, I will speak about the creation and evolution of the pipeline, and a concrete example – a day in the life of an event tracking pixel. We'll also talk about common challenges that we've overcome such as integrating different pieces of the system, schema evolution, queuing, and data retention policies.
Streaming all over the world Real life use cases with Kafka Streamsconfluent
Streaming all over the world Real life use cases with Kafka Streams, Dr. Benedikt Linse, Senior Solutions Architect, Confluent
https://www.meetup.com/Apache-Kafka-Germany-Munich/events/281819704/
Streaming data in the cloud with Confluent and MongoDB Atlas | Robert Waters,...HostedbyConfluent
Are you looking for a cloud-based architecture that includes the best of breed streaming and database technologies? In this session you will learn how to setup and configure the Confluent Cloud with MongoDB Atlas. We'll start the journey learning about the basic connectivity between the two cloud services and end with a brief discovery of what you can do with data once it is in MongoDB Atlas. By the end of this session you will know how to securely setup and configure the MongoDB Atlas connectors in the Confluent Cloud in both a source and sink configuration.
Symantec: Cassandra Data Modelling techniques in actionDataStax Academy
Our product presents an aggregated view of metadata collected for billions of objects (files, emails, sharepoint objects etc.). We used Cassandra to store those billions of objects along with aggregated view of that metadata. Customers can analyse the corpus of data in real time by searching in completely flexible way i.e. be able to get summary aggregates for many billions of objects, and then be able to further drill down to items by filtering using various facets of the metadata. We achieve this using a combination of Cassandra and ElasticSearch. This presentation will talk about various data modelling techniques we use to aggregate and then further summarise all that metadata and be able to search the summary in real t
Embracing Database Diversity with Kafka and DebeziumFrank Lyaruu
There was a time not long ago when we used relational databases for everything. Even if the data wasn’t particularly relational, we shoehorned it into relational tables, often because that was the only database we had. Thank god these dark times are over and now we have many different kinds of NoSQL databases: Document, realtime, graph, column, but that does not solve the problem that the same data might be a graph from one perspective, but a collection of documents from another.
It would be really nice if we can access that same data in many different ways, depending on the context of what we want to achieve in our current task.
As software architects this is not easy to solve but definitely possible: We can design an architecture using Event Sourcing: Capture the data with Debezium, post it to a Kafka queue, use Kafka Streams to model the data the way we like, and store the data in various different data sources, so we can synchronize data between data sources.
Bringing Streaming Data To The Masses: Lowering The “Cost Of Admission” For Y...confluent
(Bob Lehmann, Bayer) Kafka Summit SF 2018
You’ve built your streaming data platform. The early adopters are “all in” and have developed producers, consumers and stream processing apps for a number of use cases. A large percentage of the enterprise, however, has expressed interest but hasn’t made the leap. Why?
In 2014, Bayer Crop Science (formerly Monsanto) adopted a cloud first strategy and started a multi-year transition to the cloud. A Kafka-based cross-datacenter DataHub was created to facilitate this migration and to drive the shift to real-time stream processing. The DataHub has seen strong enterprise adoption and supports a myriad of use cases. Data is ingested from a wide variety of sources and the data can move effortlessly between an on premise datacenter, AWS and Google Cloud. The DataHub has evolved continuously over time to meet the current and anticipated needs of our internal customers. The “cost of admission” for the platform has been lowered dramatically over time via our DataHub Portal and technologies such as Kafka Connect, Kubernetes and Presto. Most operations are now self-service, onboarding of new data sources is relatively painless and stream processing via KSQL and other technologies is being incorporated into the core DataHub platform.
In this talk, Bob Lehmann will describe the origins and evolution of the Enterprise DataHub with an emphasis on steps that were taken to drive user adoption. Bob will also talk about integrations between the DataHub and other key data platforms at Bayer, lessons learned and the future direction for streaming data and stream processing at Bayer.
High Speed Continuous & Reliable Data Ingest into HadoopDataWorks Summit
This talk will explore the area of real-time data ingest into Hadoop and present the architectural trade-offs as well as demonstrate alternative implementations that strike the appropriate balance across the following common challenges: * Decentralized writes (multiple data centers and collectors) * Continuous Availability, High Reliability * No loss of data * Elasticity of introducing more writers * Bursts in Speed per syslog emitter * Continuous, real-time collection * Flexible Write Targets (local FS, HDFS etc.)
This is the talk I gave at the Big Data Meetup in Seattle in March. In this talk, I discuss the fundamentals of Spark Streaming and Flume, and how they integrate with each other.
Intel IT empowers business units to easily make rapid, impactful business decisions. Ingesting a variety of internal/external data sources has challenges. This slideset covers how Intel IT overcame the issues with Hadoop and Gobblin. Learn more at http://www.intel.com/itcenter
As the confluence of several mature and emerging technologies, the Internet of Things (IoT) is rapidly developing into a vibrant new marketplace. What are important considerations for technology, media, and telecom (TMT) companies as they compete for opportunities? This presentation covers:
• Questions TMT executives should be asking about impacts of IoT technologies, performance improvement opportunities, and where value can be generated.
• Building an IoT ecosystem where all players benefit – defining different players' roles and relationships, and already-successful tactics.
• Security and privacy challenges, including how data protection responsibility is assigned and monitored, and defining appropriate security and privacy standards.
Explore this quickly developing new opportunity for TMT companies.
Get more IoT insights: http://www.deloitte.com/us/iot_ecosystem
Cloudera Morphlines is a new open source framework, recently added to the CDK, that reduces the time and skills necessary to integrate, build, and change Hadoop processing applications that extract, transform, and load data into Apache Solr, Apache HBase, HDFS, enterprise data warehouses, or analytic online dashboards.
Some notes about spark streming positioning give the current players: Beam, Flink, Storm et al. Helpful if you have to choose an Streaming engine for your project.
Paris Spark Meetup Oct 26, 2015 - Spark After Dark v1.5 - Best of Advanced Ap...Chris Fregly
* Title *
Spark After Dark 1.5: Deep Dive Into Latest Perf and Scale Improvements in Spark Ecosystem
* Abstract *
Combining the most popular and technically-deep material from his wildly popular Advanced Apache Spark Meetup, Chris Fregly will provide code-level deep dives into the latest performance and scalability advancements within the Apache Spark Ecosystem by exploring the following:
1) Building a Scalable and Performant Spark SQL/DataFrames Data Source Connector such as Spark-CSV, Spark-Cassandra, Spark-ElasticSearch, and Spark-Redshift
2) Speeding Up Spark SQL Queries using Partition Pruning and Predicate Pushdowns with CSV, JSON, Parquet, Avro, and ORC
3) Tuning Spark Streaming Performance and Fault Tolerance with KafkaRDD and KinesisRDD
4) Maintaining Stability during High Scale Streaming Ingestion using Approximations and Probabilistic Data Structures from Spark, Redis, and Twitter's Algebird
5) Building Effective Machine Learning Models using Feature Engineering, Dimension Reduction, and Natural Language Processing with MLlib/GraphX, ML Pipelines, DIMSUM, Locality Sensitive Hashing, and Stanford's CoreNLP
6) Tuning Core Spark Performance by Acknowledging Mechanical Sympathy for the Physical Limitations of OS and Hardware Resources such as CPU, Memory, Network, and Disk with Project Tungsten, Asynchronous Netty, and Linux epoll
* Demos *
This talk features many interesting and audience-interactive demos - as well as code-level deep dives into many of the projects listed above.
All demo code is available on Github at the following link: https://github.com/fluxcapacitor/pipeline/wiki
In addition, the entire demo environment has been Dockerized and made available for download on Docker Hub at the following link: https://hub.docker.com/r/fluxcapacitor/pipeline/
* Speaker Bio *
Chris Fregly is a Principal Data Solutions Engineer for the newly-formed IBM Spark Technology Center, an Apache Spark Contributor, a Netflix Open Source Committer, as well as the Organizer of the global Advanced Apache Spark Meetup and Author of the Upcoming Book, Advanced Spark.
Previously, Chris was a Data Solutions Engineer at Databricks and a Streaming Data Engineer at Netflix.
When Chris isn’t contributing to Spark and other open source projects, he’s creating book chapters, slides, and demos to share knowledge with his peers at meetups and conferences throughout the world.
Workshop - How to Build Recommendation Engine using Spark 1.6 and HDP
Hands-on - Build a Data analytics application using SPARK, Hortonworks, and Zeppelin. The session explains RDD concepts, DataFrames, sqlContext, use SparkSQL for working with DataFrames and explore graphical abilities of Zeppelin.
b) Follow along - Build a Recommendation Engine - This will show how to build a predictive analytics (MLlib) recommendation engine with scoring This will give a better understanding of architecture and coding in Spark for ML.
This talk was prepared for the November 2013 DataPhilly Meetup: Data in Practice ( http://www.meetup.com/DataPhilly/events/149515412/ )
Map Reduce: Beyond Word Count by Jeff Patti
Have you ever wondered what map reduce can be used for beyond the word count example you see in all the introductory articles about map reduce? Using Python and mrjob, this talk will cover a few simple map reduce algorithms that in part power Monetate's information pipeline
Bio: Jeff Patti is a backend engineer at Monetate with a passion for algorithms, big data, and long walks on the beach. Prior to working at Monetate he performed software R&D for Lockheed Martin, where he worked on projects ranging from social network analysis to robotics.
Scala - The Simple Parts, SFScala presentationMartin Odersky
These are the slides of the talk I gave on May 22, 2014 to the San Francisco Scala user group. Similar talks were given before at GOTO Chicago, keynote, at Gilt Groupe and Hunter College in New York, at JAX Mainz and at FlatMap Oslo.
Slides presented during the Strata SF 2019 conference. Explaining how Lyft is building a multi-cluster solution for running Apache Spark on kubernetes at scale to support diverse workloads and overcome challenges.
Cloud-Native Patterns for Data-Intensive ApplicationsVMware Tanzu
Are you interested in learning how to schedule batch jobs in container runtimes?
Maybe you’re wondering how to apply continuous delivery in practice for data-intensive applications? Perhaps you’re looking for an orchestration tool for data pipelines?
Questions like these are common, so rest assured that you’re not alone.
In this webinar, we’ll cover the recent feature improvements in Spring Cloud Data Flow. More specifically, we’ll discuss data processing use cases and how they simplify the overall orchestration experience in cloud runtimes like Cloud Foundry and Kubernetes.
Please join us and be part of the community discussion!
Presenters :
Sabby Anandan, Product Manager
Mark Pollack, Software Engineer, Pivotal
GPU-Accelerating UDFs in PySpark with Numba and PyGDFKeith Kraus
With advances in computer hardware such as 10 gigabit network cards, infiniband, and solid state drives all becoming commodity offerings, the new bottleneck in big data technologies is very commonly the processing power of the CPU. In order to meet the computational demand desired by users, enterprises have had to resort to extreme scale out approaches just to get the processing power they need. One of the most well known technologies in this space, Apache Spark, has numerous enterprises publicly talking about the challenges in running multiple 1000+ node clusters to give their users the processing power they need. This talk is based on work completed by NVIDIA’s Applied Solutions Engineering team. Attendees will learn how they were able to GPU-accelerate UDFs in PySpark using open source technologies such as Numba and PyGDF, the lessons they learned in the process, and how they were able to accelerate workloads in a fraction of the hardware footprint.
CodeCamp Iasi - Creating serverless data analytics system on GCP using BigQueryMárton Kodok
Teaser: provide developers a new way of understanding advanced analytics and choosing the right cloud architecture
The new buzzword is #serverless, as there are many great services that helps us abstract away the complexity associated with managing servers. In this session we will see how serverless helps on large data analytics backends.
We will see how to architect for Cloud and implement into an existing project components that will take us into the #serverless architecture that will ingest our streaming data, run advanced analytics on petabytes of data using BigQuery on Google Cloud Platform - all this next to an existing stack, without being forced to reengineer our app.
BigQuery enables super-fast, SQL/Javascript queries against petabytes of data using the processing power of Google’s infrastructure. We will cover its core features, SQL 2011 standard, working with streaming inserts, User Defined Functions written in Javascript, reference external JS libraries, and several use cases for everyday backend developer: funnel analytics, email heatmap, custom data processing, building dashboards, extracting data using JS functions, emitting rows based on business logic.
SF Big Analytics_20190612: Scaling Apache Spark on Kubernetes at LyftChester Chen
Talk 1. Scaling Apache Spark on Kubernetes at Lyft
As part of this mission Lyft invests heavily in open source infrastructure and tooling. At Lyft Kubernetes has emerged as the next generation of cloud native infrastructure to support a wide variety of distributed workloads. Apache Spark at Lyft has evolved to solve both Machine Learning and large scale ETL workloads. By combining the flexibility of Kubernetes with the data processing power of Apache Spark, Lyft is able to drive ETL data processing to a different level. In this talk, We will talk about challenges the Lyft team faced and solutions they developed to support Apache Spark on Kubernetes in production and at scale. Topics Include: - Key traits of Apache Spark on Kubernetes. - Deep dive into Lyft's multi-cluster setup and operationality to handle petabytes of production data. - How Lyft extends and enhances Apache Spark to support capabilities such as Spark pod life cycle metrics and state management, resource prioritization, and queuing and throttling. - Dynamic job scale estimation and runtime dynamic job configuration. - How Lyft powers internal Data Scientists, Business Analysts, and Data Engineers via a multi-cluster setup.
Speaker: Li Gao
Li Gao is the tech lead in the cloud native spark compute initiative at Lyft. Prior to Lyft, Li worked at Salesforce, Fitbit, Marin Software, and a few startups etc. on various technical leadership positions on cloud native and hybrid cloud data platforms at scale. Besides Spark, Li has scaled and productionized other open source projects, such as Presto, Apache HBase, Apache Phoenix, Apache Kafka, Apache Airflow, Apache Hive, and Apache Cassandra.
Strata Singapore 2017 business use case section
"Big Telco Real-Time Network Analytics"
https://conferences.oreilly.com/strata/strata-sg/public/schedule/detail/62797
Managing Apache Spark Workload and Automatic OptimizingDatabricks
eBay is highly using Spark as one of most significant data engines. In data warehouse domain, there are millions of batch queries running every day against 6000+ key DW tables, which contains over 22PB data (compressed) and still keeps booming every year. In machine learning domain, it is playing a more and more significant role. We have introduced our great achievement in migration work from MPP database to Apache Spark last year in Europe Summit. Furthermore, from the vision of the entire infrastructure, it is still a big challenge on managing workload and efficiency for all Spark jobs upon our data center. Our team is leading the whole infrastructure of big data platform and the management tools upon it, helping our customers -- not only DW engineers and data scientists, but also AI engineers -- to leverage on the same page. In this session, we will introduce how to benefit all of them within a self-service workload management portal/system. First, we will share the basic architecture of this system to illustrate how it collects metrics from multiple data centers and how it detects the abnormal workload real-time. We develop a component called Profiler which is to enhance the current Spark core to support customized metric collection. Next, we will demonstrate some real user stories in eBay to show how the self-service system reduces the efforts both in customer side and infra-team side. That's the highlight part about Spark job analysis and diagnosis. Finally, some incoming advanced features will be introduced to describe an automatic optimizing workflow rather than just alerting.
Speaker: Lantao Jin
Webinar: Enterprise Data Management in the Era of MongoDB and Data LakesMongoDB
With so much talk of how Big Data is revolutionizing the world and how a data lake with Hadoop and/or Spark will solve all your data problems, it is hard to tell what is hype, reality, or somewhere in-between.
In working with dozens of enterprises in varying stages of their enterprise data management (EDM) strategy, MongoDB enterprise architect, Matt Kalan, sees the same challenges and misunderstandings arise again and again.
In this session, he will explain common challenges in data management, what capabilities are necessary, and what the future state of architecture looks like. MongoDB is uniquely capable of filling common gaps in the data lake strategy.
This session also includes a live Q&A portion during which you are encouraged to ask questions of our team.
Scaling Apache Spark on Kubernetes at LyftDatabricks
Lyft is on the mission to improve people's lives with the world's best transportation. As part of this mission Lyft invests heavily in open source infrastructure and tooling. At Lyft Kubernetes has emerged as the next generation of cloud native infrastructure to support a wide variety of distributed workloads. Apache Spark at Lyft has evolved to solve both Machine Learning and large scale ETL workloads. By combining the flexibility of Kubernetes with the data processing power of Apache Spark, Lyft is able to drive ETL data processing to a different level. In this talk, Li Gao and Rohit Menon will talk about challenges the Lyft team faced and solutions they developed to support Apache Spark on Kubernetes in production and at scale. Topics Include: - Key traits of Apache Spark on Kubernetes. - Deep dive into Lyft's multi-cluster setup and operationality to handle petabytes of production data. - How Lyft extends and enhances Apache Spark to support capabilities such as Spark pod life cycle metrics and state management, resource prioritization, and queuing and throttling. - Dynamic job scale estimation and runtime dynamic job configuration. - How Lyft powers internal Data Scientists, Business Analysts, and Data Engineers via a multi-cluster setup.
Speakers: Li Gao, Rohit Menon
Why Wait? Realtime Ingestion With Chen Qin and Heng Zhang | Current 2022HostedbyConfluent
Historically, Pinterest data warehouse ingestion and indexing services were implemented on batch ETL and Kafka streaming respectively. As the product side leans more toward real-time and near-realtime data to innovate and compete, teams work together to revamp the ingestion and processing stack in Pinterest.
In this talk, we plan to share our near-real-time ingestion system built on top of Apache Kafka, Apache Flink, and Apache Iceberg. We pick ANSI SQL as the common currency to minimize the ""lambda architecture"" learning curve of teams adopting fresh data near-realtime data.
InfoSphere BigInsights - Analytics power for Hadoop - field experienceWilfried Hoge
How to analyze binary data as a technical business user. Use InfoSphere BigInsights to bring analytics on Hadoop closer to a user.
Presented at the OOP conference in Munich, 27.01.2015
Originally presented at Strata EU 2017: https://conferences.oreilly.com/strata/strata-eu/public/schedule/detail/57631
Cloud providers currently offer convenient on-demand managed big data clusters (PaaS) with a pay-as-you-go model. In PaaS, analytical engines such as Spark and Hive come ready to use, with a general-purpose configuration and upgrade management. Over the last year, the Spark framework and APIs have been evolving very rapidly, with major improvements on performance and the release of v2, making it challenging to keep up-to-date production services both on-premises and in the cloud for compatibility and stability.
Nicolas Poggi evaluates the out-of-the-box support for Spark and compares the offerings, reliability, scalability, and price-performance from major PaaS providers, including Azure HDinsight, Amazon Web Services EMR, Google Dataproc, and Rackspace Cloud Big Data, with an on-premises commodity cluster as baseline. Nicolas uses BigBench, the brand new standard (TPCx-BB) for big data systems, with both Spark and Hive implementations for benchmarking the systems. BigBench combines SQL queries, MapReduce, user code (UDF), and machine learning, which makes it ideal to stress Spark libraries (SparkSQL, DataFrames, MLlib, etc.).
The work is framed within the ALOJA research project, which features an open source benchmarking and analysis platform that has been recently extended to support SQL-on-Hadoop engines and BigBench. The ALOJA project aims to lower the total cost of ownership (TCO) of big data deployments and study their performance characteristics for optimization. Nicolas highlights how to easily repeat the benchmarks through ALOJA and benefit from BigBench to optimize your Spark cluster for advanced users. The work is a continuation of a paper to be published at the IEEE Big Data 16 conference. (A preprint copy can be obtained here.)
Similar to Gobblin: Unifying Data Ingestion for Hadoop (20)
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.
Adjusting primitives for graph : SHORT REPORT / NOTESSubhajit Sahu
Graph algorithms, like PageRank Compressed Sparse Row (CSR) is an adjacency-list based graph representation that is
Multiply with different modes (map)
1. Performance of sequential execution based vs OpenMP based vector multiply.
2. Comparing various launch configs for CUDA based vector multiply.
Sum with different storage types (reduce)
1. Performance of vector element sum using float vs bfloat16 as the storage type.
Sum with different modes (reduce)
1. Performance of sequential execution based vs OpenMP based vector element sum.
2. Performance of memcpy vs in-place based CUDA based vector element sum.
3. Comparing various launch configs for CUDA based vector element sum (memcpy).
4. Comparing various launch configs for CUDA based vector element sum (in-place).
Sum with in-place strategies of CUDA mode (reduce)
1. Comparing various launch configs for CUDA based vector element sum (in-place).
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.”
10. (18)
Case Study – Filtering Sensitive Data
9
Has Sensitive
Data?
no
Source
Extractor
WorkUnit
Converter and
Quality Checker
Fork and Branching
Writer
DataPublisher
Writer
Sensitive Data
Filtering Converter
yes
12. (18)
State and Metadata Mgmt.
11
State Store
- Stores runtime metadata, e.g., checkpoints
(a.k.a. watermarks)
~ Carried over between job runs
- Default impl: serializes job/task states into
files, one per run.
- Allows other implementations that conform
to the interface to be plugged in.
State Store
job run #2
job run #3job run #1
SEP
2
SEP
3
SEP
2 SEP
3
EXAMPLE
14. (18)
Running Modes
13
Standalone
Runs in a single
JVM; tasks run in a
thread pool.
Scale-out with
MapReduce
Each job run launches
a MR job, using
mappers as containers
to run tasks.
Scale-out with
General
Distributed
Resource Manager
Supports long-running
continuous ingestion,
with better resource
utilization and SLA
guarantees.
YARN
*in progress
15. (18)
Gobblin in Production @ LinkedIn
• In production since 2014
• Usages
– Internal sources HDFS
• Kafka, MySQL, Dropbox, etc.
– External sources HDFS
• Salesforce, GoogleAnalytics, S3, etc.
– HDFS HDFS
• Closed member data purging
– Egress from HDFS (future work)
• Data volume
– Over a dozen data sources,
– thousands of datasets,
– tens ofTBs,
… daily.
14