Alternatives to Amazon EMR

To provide employees with the critical need of interactive querying, we’ve worked with Presto, an open-source distributed SQL query engine, over the years. Operating Presto at Pinterest’s scale has involved resolving quite a few challenges like, supporting deeply nested and huge thrift schemas, slow/ bad worker detection and remediation, auto-scaling cluster, graceful cluster shutdown and impersonation support for ldap authenticator.

Our infrastructure is built on top of Amazon EC2 and we leverage Amazon S3 for storing our data. This separates compute and storage layers, and allows multiple compute clusters to share the S3 data.

We have hundreds of petabytes of data and tens of thousands of Apache Hive tables. Our Presto clusters are comprised of a fleet of 450 r4.8xl EC2 instances. Presto clusters together have over 100 TBs of memory and 14K vcpu cores. Within Pinterest, we have close to more than 1,000 monthly active users (out of total 1,600+ Pinterest employees) using Presto, who run about 400K queries on these clusters per month.

Each query submitted to Presto cluster is logged to a Kafka topic via Singer. Singer is a logging agent built at Pinterest and we talked about it in a previous post. Each query is logged when it is submitted and when it finishes. When a Presto cluster crashes, we will have query submitted events without corresponding query finished events. These events enable us to capture the effect of cluster crashes over time.

Each Presto cluster at Pinterest has workers on a mix of dedicated AWS EC2 instances and Kubernetes pods. Kubernetes platform provides us with the capability to add and remove workers from a Presto cluster very quickly. The best-case latency on bringing up a new worker on Kubernetes is less than a minute. However, when the Kubernetes cluster itself is out of resources and needs to scale up, it can take up to ten minutes. Some other advantages of deploying on Kubernetes platform is that our Presto deployment becomes agnostic of cloud vendor, instance types, OS, etc.

#BigData #AWS #DataScience #DataEngineering

Our whole DevOps stack consists of the following tools:

• GitHub (incl. GitHub Pages/Markdown for Documentation, GettingStarted and HowTo's) for collaborative review and code management tool
• Respectively Git as revision control system
• SourceTree as Git GUI
• Visual Studio Code as IDE
• CircleCI for continuous integration (automatize development process)
• Prettier / TSLint / ESLint as code linter
• SonarQube as quality gate
• Docker as container management (incl. Docker Compose for multi-container application management)
• VirtualBox for operating system simulation tests
• Kubernetes as cluster management for docker containers
• Heroku for deploying in test environments
• nginx as web server (preferably used as facade server in production environment)
• SSLMate (using OpenSSL) for certificate management
• Amazon EC2 (incl. Amazon S3) for deploying in stage (production-like) and production environments
• PostgreSQL as preferred database system
• Redis as preferred in-memory database/store (great for caching)

The main reason we have chosen Kubernetes over Docker Swarm is related to the following artifacts:

• Key features: Easy and flexible installation, Clear dashboard, Great scaling operations, Monitoring is an integral part, Great load balancing concepts, Monitors the condition and ensures compensation in the event of failure.
• Applications: An application can be deployed using a combination of pods, deployments, and services (or micro-services).
• Functionality: Kubernetes as a complex installation and setup process, but it not as limited as Docker Swarm.
• Monitoring: It supports multiple versions of logging and monitoring when the services are deployed within the cluster (Elasticsearch/Kibana (ELK), Heapster/Grafana, Sysdig cloud integration).
• Scalability: All-in-one framework for distributed systems.
• Other Benefits: Kubernetes is backed by the Cloud Native Computing Foundation (CNCF), huge community among container orchestration tools, it is an open source and modular tool that works with any OS.

The early data ingestion pipeline at Pinterest used Kafka as the central message transporter, with the app servers writing messages directly to Kafka, which then uploaded log files to S3.

For databases, a custom Hadoop streamer pulled database data and wrote it to S3.

Challenges cited for this infrastructure included high operational overhead, as well as potential data loss occurring when Kafka broker outages led to an overflow of in-memory message buffering.

Why we built Marmaray, an open source generic data ingestion and dispersal framework and library for Apache Hadoop :

Built and designed by our Hadoop Platform team, Marmaray is a plug-in-based framework built on top of the Hadoop ecosystem. Users can add support to ingest data from any source and disperse to any sink leveraging the use of Apache Spark . The name, Marmaray, comes from a tunnel in Turkey connecting Europe and Asia. Similarly, we envisioned Marmaray within Uber as a pipeline connecting data from any source to any sink depending on customer preference:

https://eng.uber.com/marmaray-hadoop-ingestion-open-source/

(Direct GitHub repo: https://github.com/uber/marmaray Kafka Kafka Manager )

We are in the process of building a modern content platform to deliver our content through various channels. We decided to go with Microservices architecture as we wanted scale. Microservice architecture style is an approach to developing an application as a suite of small independently deployable services built around specific business capabilities. You can gain modularity, extensive parallelism and cost-effective scaling by deploying services across many distributed servers. Microservices modularity facilitates independent updates/deployments, and helps to avoid single point of failure, which can help prevent large-scale outages. We also decided to use Event Driven Architecture pattern which is a popular distributed asynchronous architecture pattern used to produce highly scalable applications. The event-driven architecture is made up of highly decoupled, single-purpose event processing components that asynchronously receive and process events.

To build our #Backend capabilities we decided to use the following: 1. #Microservices - Java with Spring Boot , Node.js with ExpressJS and Python with Flask 2. #Eventsourcingframework - Amazon Kinesis , Amazon Kinesis Firehose , Amazon SNS , Amazon SQS, AWS Lambda 3. #Data - Amazon RDS , Amazon DynamoDB , Amazon S3 , MongoDB Atlas

To build #Webapps we decided to use Angular 2 with RxJS

#Devops - GitHub , Travis CI , Terraform , Docker , Serverless

Back in 2014, I was given an opportunity to re-architect SmartZip Analytics platform, and flagship product: SmartTargeting. This is a SaaS software helping real estate professionals keeping up with their prospects and leads in a given neighborhood/territory, finding out (thanks to predictive analytics) who's the most likely to list/sell their home, and running cross-channel marketing automation against them: direct mail, online ads, email... The company also does provide Data APIs to Enterprise customers.

I had inherited years and years of technical debt and I knew things had to change radically. The first enabler to this was to make use of the cloud and go with AWS, so we would stop re-inventing the wheel, and build around managed/scalable services.

For the SaaS product, we kept on working with Rails as this was what my team had the most knowledge in. We've however broken up the monolith and decoupled the front-end application from the backend thanks to the use of Rails API so we'd get independently scalable micro-services from now on.

Our various applications could now be deployed using AWS Elastic Beanstalk so we wouldn't waste any more efforts writing time-consuming Capistrano deployment scripts for instance. Combined with Docker so our application would run within its own container, independently from the underlying host configuration.

Storage-wise, we went with Amazon S3 and ditched any pre-existing local or network storage people used to deal with in our legacy systems. On the database side: Amazon RDS / MySQL initially. Ultimately migrated to Amazon RDS for Aurora / MySQL when it got released. Once again, here you need a managed service your cloud provider handles for you.

Future improvements / technology decisions included:

Caching: Amazon ElastiCache / Memcached CDN: Amazon CloudFront Systems Integration: Segment / Zapier Data-warehousing: Amazon Redshift BI: Amazon Quicksight / Superset Search: Elasticsearch / Amazon Elasticsearch Service / Algolia Monitoring: New Relic

As our usage grows, patterns changed, and/or our business needs evolved, my role as Engineering Manager then Director of Engineering was also to ensure my team kept on learning and innovating, while delivering on business value.

One of these innovations was to get ourselves into Serverless : Adopting AWS Lambda was a big step forward. At the time, only available for Node.js (Not Ruby ) but a great way to handle cost efficiency, unpredictable traffic, sudden bursts of traffic... Ultimately you want the whole chain of services involved in a call to be serverless, and that's when we've started leveraging Amazon DynamoDB on these projects so they'd be fully scalable.

Back in 2014, I was given an opportunity to re-architect SmartZip Analytics platform, and flagship product: SmartTargeting. This is a SaaS software helping real estate professionals keeping up with their prospects and leads in a given neighborhood/territory, finding out (thanks to predictive analytics) who's the most likely to list/sell their home, and running cross-channel marketing automation against them: direct mail, online ads, email... The company also does provide Data APIs to Enterprise customers.

I had inherited years and years of technical debt and I knew things had to change radically. The first enabler to this was to make use of the cloud and go with AWS, so we would stop re-inventing the wheel, and build around managed/scalable services.

For the SaaS product, we kept on working with Rails as this was what my team had the most knowledge in. We've however broken up the monolith and decoupled the front-end application from the backend thanks to the use of Rails API so we'd get independently scalable micro-services from now on.

Our various applications could now be deployed using AWS Elastic Beanstalk so we wouldn't waste any more efforts writing time-consuming Capistrano deployment scripts for instance. Combined with Docker so our application would run within its own container, independently from the underlying host configuration.

Storage-wise, we went with Amazon S3 and ditched any pre-existing local or network storage people used to deal with in our legacy systems. On the database side: Amazon RDS / MySQL initially. Ultimately migrated to Amazon RDS for Aurora / MySQL when it got released. Once again, here you need a managed service your cloud provider handles for you.

Future improvements / technology decisions included:

Caching: Amazon ElastiCache / Memcached CDN: Amazon CloudFront Systems Integration: Segment / Zapier Data-warehousing: Amazon Redshift BI: Amazon Quicksight / Superset Search: Elasticsearch / Amazon Elasticsearch Service / Algolia Monitoring: New Relic

As our usage grows, patterns changed, and/or our business needs evolved, my role as Engineering Manager then Director of Engineering was also to ensure my team kept on learning and innovating, while delivering on business value.

One of these innovations was to get ourselves into Serverless : Adopting AWS Lambda was a big step forward. At the time, only available for Node.js (Not Ruby ) but a great way to handle cost efficiency, unpredictable traffic, sudden bursts of traffic... Ultimately you want the whole chain of services involved in a call to be serverless, and that's when we've started leveraging Amazon DynamoDB on these projects so they'd be fully scalable.

Looker , Stitch , Amazon Redshift , dbt

We recently moved our Data Analytics and Business Intelligence tooling to Looker . It's already helping us create a solid process for reusable SQL-based data modeling, with consistent definitions across the entire organizations. Looker allows us to collaboratively build these version-controlled models and push the limits of what we've traditionally been able to accomplish with analytics with a lean team.

For Data Engineering, we're in the process of moving from maintaining our own ETL pipelines on AWS to a managed ELT system on Stitch. We're also evaluating the command line tool, dbt to manage data transformations. Our hope is that Stitch + dbt will streamline the ELT bit, allowing us to focus our energies on analyzing data, rather than managing it.

From my point of view, both OpenRefine and Apache Hive serve completely different purposes. OpenRefine is intended for interactive cleaning of messy data locally. You could work with their libraries to use some of OpenRefine features as part of your data pipeline (there are pointers in FAQ), but OpenRefine in general is intended for a single-user local operation.

I can't recommend a particular alternative without better understanding of your use case. But if you are looking for an interactive tool to work with big data at scale, take a look at notebook environments like Jupyter, Databricks, or Deepnote. If you are building a data processing pipeline, consider also Apache Spark.

Edit: Fixed references from Hadoop to Hive, which is actually closer to Spark.

The algorithms and data infrastructure at Stitch Fix is housed in #AWS. Data acquisition is split between events flowing through Kafka, and periodic snapshots of PostgreSQL DBs. We store data in an Amazon S3 based data warehouse. Apache Spark on Yarn is our tool of choice for data movement and #ETL. Because our storage layer (s3) is decoupled from our processing layer, we are able to scale our compute environment very elastically. We have several semi-permanent, autoscaling Yarn clusters running to serve our data processing needs. While the bulk of our compute infrastructure is dedicated to algorithmic processing, we also implemented Presto for adhoc queries and dashboards.

Beyond data movement and ETL, most #ML centric jobs (e.g. model training and execution) run in a similarly elastic environment as containers running Python and R code on Amazon EC2 Container Service clusters. The execution of batch jobs on top of ECS is managed by Flotilla, a service we built in house and open sourced (see https://github.com/stitchfix/flotilla-os).

At Stitch Fix, algorithmic integrations are pervasive across the business. We have dozens of data products actively integrated systems. That requires serving layer that is robust, agile, flexible, and allows for self-service. Models produced on Flotilla are packaged for deployment in production using Khan, another framework we've developed internally. Khan provides our data scientists the ability to quickly productionize those models they've developed with open source frameworks in Python 3 (e.g. PyTorch, sklearn), by automatically packaging them as Docker containers and deploying to Amazon ECS. This provides our data scientist a one-click method of getting from their algorithms to production. We then integrate those deployments into a service mesh, which allows us to A/B test various implementations in our product.

For more info:

• Our Algorithms Tour: https://algorithms-tour.stitchfix.com/
• Our blog: https://multithreaded.stitchfix.com/blog/
• Careers: https://multithreaded.stitchfix.com/careers/

#DataScience #DataStack #Data

As a frontend engineer on the Algorithms & Analytics team at Stitch Fix, I work with data scientists to develop applications and visualizations to help our internal business partners make data-driven decisions. I envisioned a platform that would assist data scientists in the data exploration process, allowing them to visually explore and rapidly iterate through their assumptions, then share their insights with others. This would align with our team's philosophy of having engineers "deploy platforms, services, abstractions, and frameworks that allow the data scientists to conceive of, develop, and deploy their ideas with autonomy", and solve the pain of data exploration.

The final product, code-named Dora, is built with React, Redux.js and Victory, backed by Elasticsearch to enable fast and iterative data exploration, and uses Apache Spark to move data from our Amazon S3 data warehouse into the Elasticsearch cluster.

I'm wondering if any Cloud Firestore users might be open to sharing some input and challenges encountered when trying to create a low-cost, low-latency data pipeline to their Analytics warehouse (e.g. Google BigQuery, Snowflake, etc...)

I'm working with a platform by the name of Estuary.dev, an ETL/ELT and we are conducting some research on the pain points here to see if there are drawbacks of the Firestore->BQ extension and/or if users are seeking easy ways for getting nosql->fine-grained tabular data

Please feel free to drop some knowledge/wish list stuff on me for a better pipeline here!

I use Google BigQuery because it makes is super easy to query and store data for analytics workloads. If you're using GCP, you're likely using BigQuery. However, running data viz tools directly connected to BigQuery will run pretty slow. They recently announced BI Engine which will hopefully compete well against big players like Snowflake when it comes to concurrency.

What's nice too is that it has SQL-based ML tools, and it has great GIS support!