Distributed Deep Learning with Docker at Salesforce
0 likes482 views
The document outlines how Salesforce employs Docker for distributed deep learning model deployment, addressing challenges such as team specialization, interaction with model servers, and testing. It emphasizes the importance of preprocessing, inference, and postprocessing in the deep learning workflow and presents solutions like containerizing functions and using Docker Compose for testing. Key takeaways include the scalability of deep learning processes through Docker and its ability to simplify complex systems.
1 of 28
Downloaded 17 times
![Deep Learning at Salesforce
Discusses
cat
preprocess
[0.2, 0.71, 0.89, 0.6]
[0.85, 0.15]
inference
postprocess“Hello! My
cat is
friendly.”](https://image.slidesharecdn.com/jeffhajewski-distributeddeeplearning-200529022235/85/Distributed-Deep-Learning-with-Docker-at-Salesforce-9-320.jpg)
![Deep Learning at Salesforce
Discusses
cat
preprocess
[0.2, 0.71, 0.89, 0.6]
[0.85, 0.15]
inference
postprocess“Hello! My
cat is
friendly.”
Data Science Team](https://image.slidesharecdn.com/jeffhajewski-distributeddeeplearning-200529022235/85/Distributed-Deep-Learning-with-Docker-at-Salesforce-10-320.jpg)
![Deep Learning at Salesforce
Discusses
cat
preprocess
[0.2, 0.71, 0.89, 0.6]
[0.85, 0.15]
inference
postprocess“Hello! My
cat is
friendly.”
Data Science Team Systems Team](https://image.slidesharecdn.com/jeffhajewski-distributeddeeplearning-200529022235/85/Distributed-Deep-Learning-with-Docker-at-Salesforce-11-320.jpg)
Ad
Recommended
Monolithic to Microservices + Docker = SDLC on Steroids!
Monolithic to Microservices + Docker = SDLC on Steroids!Docker, Inc. The document discusses the transition from monolithic architecture to microservices, highlighting the adoption of Docker and a polyglot tech stack at SS&C Eze to enhance the software development lifecycle. It emphasizes improvements in developer experience, quick onboarding, and the effectiveness of testing and deployment strategies, such as branch by abstraction and feature toggles. Key takeaways include the necessity for strict engineering discipline and the advantages of testing services in isolation for more efficient delivery.
DCSF19 Containerized Databases for Enterprise Applications
DCSF19 Containerized Databases for Enterprise ApplicationsDocker, Inc. The document discusses the role of containerized databases in managing mission-critical enterprise applications, emphasizing the importance of data persistence and security. It outlines misconceptions about container databases, highlighting the benefits of immutability and portability, while urging smart design for data management. Key practices for enhancing data security, such as building trusted images and ensuring secure access, are also presented.
DCSF19 Transforming a 15+ Year Old Semiconductor Manufacturing Environment
DCSF19 Transforming a 15+ Year Old Semiconductor Manufacturing EnvironmentDocker, Inc. The document discusses a transformation initiative at Jireh Semiconductor, led by Jeanie M. Schwenk, aimed at modernizing software systems and transitioning from a waterfall methodology to Agile Scrum. The plan includes coaching the team, modernizing hardware and operating systems, and utilizing containerization to improve application deployment while minimizing manufacturing downtime. Throughout the process, challenges related to legacy systems, team resistance to change, and technical hurdles in virtualization and dependency management are addressed.
Infrastructure as Code with Ansible
Infrastructure as Code with AnsibleDaniel Bezerra This document discusses Infrastructure as Code (IaC) and how Ansible can be used to implement IaC. It defines IaC, explains the benefits such as cost reduction, speed, and risk removal. It also covers IaC approaches, methods, idempotence, best practices, and provides an Ansible demo.
Monitoring in a Microservices World
Monitoring in a Microservices WorldDocker, Inc. Docker facilitates the creation of dynamic, distributed, and scalable polyglot applications but necessitates a novel monitoring approach. Key strategies include extensive automation, utilization of metrics and distributed traces, and adapting organizational roles such as DevOps and SRE to evolving responsibilities. Challenges include managing dynamically changing environments and ensuring comprehensive application state representation.
Puppet overview
Puppet overviewjoshbeard The document outlines Puppet Enterprise and its role in IT automation, emphasizing automation practices for high-performing organizations. It describes how Puppet manages infrastructure through configuration, orchestration, and reporting, helping streamline operations while ensuring desired states across environments. Additionally, it highlights the community growth and various functionalities offered by Puppet, including support for cloud provisioning and role-based access control.
Database deployments - dotnetsheff
Database deployments - dotnetsheffGiulio Vian The document discusses database deployment strategies, contrasting migration-based and state-based approaches while emphasizing the importance of handling conflicts and dependencies during deployments. It highlights tools like SSDT, Redgate SQL Compare, Flyway, and DBUP, providing references for further reading. The author, Giulio Vian, underscores that careful planning and knowledge are crucial to navigate the complexities of database refactoring.
DockerCon 16 General Session Day 1
DockerCon 16 General Session Day 1Docker, Inc. This document discusses Docker and its goals of making development and operations tools easier to use. It outlines Docker's focus on three areas: 1) improving the developer experience by creating tools that are simple and intuitive, 2) building orchestration capabilities directly into Docker to simplify managing multiple containers, and 3) enhancing the operations experience through deep integration with cloud platforms and a new application bundle format. The document promotes Docker 1.12 which includes built-in orchestration functionality and announces public betas of Docker for Mac and Windows.
Predicting Space Weather with Docker
Predicting Space Weather with DockerDocker, Inc. The document discusses how NOAA's Space Weather Prediction Center transitioned from a monolithic architecture to microservices using Docker. It describes how they started with a small verification project, then replaced their critical GOES satellite data source. This improved developers' morale and delivery speed. They encountered some security issues initially but learned from them. The transition was very successful and allowed them to quickly expand their mission to forecast aviation impacts using scientists' models packaged as Docker services.
DCSF19 Container Security: Theory & Practice at Netflix
DCSF19 Container Security: Theory & Practice at NetflixDocker, Inc. Michael Wardrop discusses container security at Netflix, emphasizing the importance of tailored security solutions based on individual threat models. The document outlines the evolution of Netflix's container management platform, Titus, and highlights various security practices, including identity management, mutual TLS, and vulnerability management strategies. Key takeaways stress the significance of protecting cloud control planes, the necessity of involving security in the development lifecycle, and the importance of providing engineers with the tools to debug and monitor container performance.
Hands-on Helm
Hands-on Helm Docker, Inc. The document discusses Helm, a package manager for Kubernetes, providing an overview of its features and significant changes in the transition from Helm 2 to Helm 3. Key updates include improved security, simplified architecture, and modifications to CLI commands and release metadata management. The document also encourages community involvement and outlines considerations for migration and support of existing charts.
Using the SDACK Architecture on Security Event Inspection by Yu-Lun Chen and ...
Using the SDACK Architecture on Security Event Inspection by Yu-Lun Chen and ...Docker, Inc. The document presents insights on using the SDACK architecture for security event inspection, emphasizing the need for scalable, on-premises solutions to handle large data volumes and improve actionable insights from logs. It outlines the benefits of dockerizing applications for seamless deployment, testing, and scaling, along with considerations for security and configuration management. Key lessons learned highlight the importance of flexibility, reproducibility, and effective monitoring to enhance the development and operation of security-monitoring systems.
Puppet plugin for vRealize Automation (vRA)
Puppet plugin for vRealize Automation (vRA)Puppet The document discusses the integration of Puppet Enterprise and vRealize Automation (vRA) to enhance digital transformation, DevOps initiatives, and cloud migration. It highlights the benefits of automation, including increased deployment speed, reduced outages, and improved security and compliance measures. The collaboration aims to modernize IT processes, eliminate manual workflows, and provide a unified management platform for seamless operations.
Simple tweaks to get the most out of your JVM
Simple tweaks to get the most out of your JVMJamie Coleman The document discusses the Java Virtual Machine (JVM), its importance, architecture, and how cloud computing has influenced its use and optimization. It covers different JVM implementations including Hotspot, OpenJ9, and GraalVM, providing insights into their features and use cases. Additionally, it emphasizes performance tuning, efficient coding practices, and the balance of metrics for optimizing JVM performance in cloud environments.
Webinar: Accelerate Your Inner Dev Loop for Kubernetes Services
Webinar: Accelerate Your Inner Dev Loop for Kubernetes Services Ambassador Labs The document discusses an approach called Service Preview that allows developers to accelerate their inner development loops for Kubernetes services. Service Preview allows developers to code microservices locally, preview changes immediately with routed test traffic, use local tools like debuggers while treating the microservice as if it is live in the cluster. This is done in isolation before merging changes. Service Preview is presented as a simple way to remedy the complexities introduced to development workflows by containers and Kubernetes without the drawbacks of other approaches.
DCSF 19 How Entergy is Mitigating Legacy Windows Operating System Vulnerabili...
DCSF 19 How Entergy is Mitigating Legacy Windows Operating System Vulnerabili...Docker, Inc. The document outlines a large-scale IT security initiative by Entergy Corporation to mitigate vulnerabilities in legacy Windows operating systems using Docker Enterprise. It discusses the benefits of containerization, including reduced maintenance overhead and licensing costs, while integrating DevSecOps practices. Key lessons learned emphasize the need for transformation in large enterprises and the importance of project management and cultural change during this process.
DCSF 19 Microservices API: Routing Across Any Infrastructure
DCSF 19 Microservices API: Routing Across Any InfrastructureDocker, Inc. The document discusses the challenges of traditional infrastructure and the need for modern solutions such as microservices and API servers. It highlights issues with scaling, deployment methods, and routing across infrastructures while proposing solutions like orchestration with Kubernetes and Swarm. The focus is on achieving self-service redundancy, availability, and effective rate limiting in a production environment.
DCSF 19 Developing Apps with Containers, Functions and Cloud Services
DCSF 19 Developing Apps with Containers, Functions and Cloud ServicesDocker, Inc. Patrick Chanezon, a cloud advocate at Microsoft, discusses developing applications using containers, functions, and managed cloud services, emphasizing Microsoft's mission to empower organizations. He elaborates on cloud app abstractions including Docker for containers and Azure Functions for serverless computing, highlighting their features like scalability and event-driven capabilities. The document also covers development workflows, Kubernetes integration, and recent innovations in cloud-native application management like the Cloud Native Application Bundles (CNAB).
Serverless java
Serverless javaVishwas N 1. The document discusses serverless computing and its advantages over traditional server-based systems like scalability and reduced costs.
2. It notes some potential downsides of serverless like difficulties testing and debugging as well as security and vendor lock-in concerns.
3. The document provides an overview of serverless concepts like APIs, operations, scripting and functions and compares the serverless model to Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) models.
DCSF 19 Modernizing Insurance with Docker Enterprise: The Physicians Mutual ...
DCSF 19 Modernizing Insurance with Docker Enterprise: The Physicians Mutual ...Docker, Inc. The document outlines the modernization journey of Physicians Mutual's application development, detailing the transition from legacy systems to contemporary practices such as RESTful APIs and containerization using Docker. It emphasizes the importance of intentional collaboration and addressing stakeholder concerns to maintain momentum during the initiative. Outcomes include the establishment of numerous services and containers, improved high availability, and enhanced monitoring capabilities.
DCEU 18: From Monolith to Microservices
DCEU 18: From Monolith to MicroservicesDocker, Inc. The document discusses the challenges of transitioning from monolithic architectures to microservices, emphasizing release engineering, independent component lifecycles, and the impact of organizational structures on design. It outlines various tools and strategies for achieving operational excellence and managing complexity in microservices, including contract validation, dependency management, and continuous integration testing. Additionally, it highlights the significance of monitoring and health checks in distinguishing between single service and multiple service environments.
Accessible hpc for everyone with docker and containers
Accessible hpc for everyone with docker and containersDocker, Inc. The document discusses the advancements and challenges in high-performance computing (HPC), particularly with the integration of Docker and container technology. It highlights the evolution of HPC systems over decades and explains how containers can address barriers such as cost and expertise for scientists. Furthermore, it outlines current HPC scheduling methods, integration of AI, and opportunities for collaboration in the HPC community through various upcoming conferences.
DCEU 18: 5 Patterns for Success in Application Transformation
DCEU 18: 5 Patterns for Success in Application TransformationDocker, Inc. The document outlines five successful patterns for transforming Docker applications, emphasizing logging, configuration, dependencies, health checks, and metrics. It provides insights into how to enhance old applications without altering code or disrupting development workflows. Additionally, it includes links for further information on Docker trials and video resources for .NET and Java.
Networking in Docker EE 2.0 with Kubernetes and Swarm
Networking in Docker EE 2.0 with Kubernetes and SwarmAbhinandan P.b The document discusses Docker EE 2.0's networking capabilities with Kubernetes and Swarm, focusing on security, multi-tenancy, and observability. It highlights design principles, deployment options, and the flexibility of network configurations using various drivers. The conclusion emphasizes the choice between Swarm for simplicity and Kubernetes for greater flexibility and cloud integration.
Application Deployment and Management at Scale with 1&1 by Matt Baldwin
Application Deployment and Management at Scale with 1&1 by Matt BaldwinDocker, Inc. 1&1 manages application deployment and infrastructure for customers in a scalable, multi-tenant environment, addressing challenges of a monolithic architecture by moving towards a container-based model. The new architecture allows better isolation, independent scalability, and reduces management overhead while providing users with more customization options. Future goals include improved resource management, automated lifecycle management, and enhanced compatibility with various application components.
JEEconf 2017
JEEconf 2017Ihor Kolodyuk This document discusses automated scaling of JavaEE microservice stacks. It begins by introducing the speaker and their experience with autoscaling technologies. It then explains that while simple applications can be easily scaled by increasing container replicas, scaling complex JavaEE applications requires more work. The document outlines three key aspects needed for scalable applications: using lightweight application servers adapted for containers; decomposing monolithic applications into independently scalable microservices; and using orchestration software to automate provisioning, health checks, metrics collection, and scaling. It provides examples of scaling strategies for different application components like message queues and databases. It also discusses common issues like ensuring new instances are properly registered for events and services. The document emphasizes that proper configuration of autoscaling triggers
Container on azure
Container on azureVishwas N This document discusses containers and container orchestration on Azure. It begins with an introduction to containers and their advantages over virtual machines. It then covers building Dockerfiles, container commands, and hosting container registries and applications on Azure. Container orchestration with Kubernetes is discussed as a way to deploy and scale containerized applications on the cloud, providing capabilities like auto-scaling, self-healing, service discovery and load balancing. The document points to additional future content on using Azure Kubernetes Service.
Immutable Awesomeness by John Willis and Josh Corman
Immutable Awesomeness by John Willis and Josh CormanDocker, Inc. The document discusses the intersection of DevOps and immutable infrastructure, highlighting the importance of high performance organizational capital and the challenges in supply chain management. It analyzes various software vulnerabilities and approaches to ensure consistent and reliable infrastructure deployment, advocating for immutable delivery and principles like variety, velocity, variability, and visibility. Additionally, it references several materials and methodologies that can enhance agility and efficiency in DevOps practices.
No BS Guide to Deep Learning in the Enterprise
No BS Guide to Deep Learning in the EnterpriseJesus Rodriguez The document provides a comprehensive overview of deep learning in the enterprise, highlighting its challenges, methodologies, and current frameworks. It discusses various deep learning stack categories, including tiered frameworks and cloud solutions, emphasizing the complexity of implementation and the required expertise. The summary underscores that despite the crowded market, successful deep learning applications should start small and iterate quickly.
Why scala for data science
Why scala for data scienceGuglielmo Iozzia This document discusses why Scala is a good programming language for data science. It begins by providing background on Scala as a functional programming language that runs on the Java Virtual Machine. The main reasons given for using Scala in data science are its robustness for large datasets, integration with common big data tools that run on JVM, and available libraries like Spark MLlib, DeepLearning4J, and ND4J. Code examples are provided showing how to perform tasks with these libraries in Scala. The document also discusses how Scala, Python, and Keras can be used together via TensorFlow for prototyping models in Python and deploying them in Scala applications using DeepLearning4J.
More Related Content
What's hot (20)
Predicting Space Weather with Docker
Predicting Space Weather with DockerDocker, Inc. The document discusses how NOAA's Space Weather Prediction Center transitioned from a monolithic architecture to microservices using Docker. It describes how they started with a small verification project, then replaced their critical GOES satellite data source. This improved developers' morale and delivery speed. They encountered some security issues initially but learned from them. The transition was very successful and allowed them to quickly expand their mission to forecast aviation impacts using scientists' models packaged as Docker services.
DCSF19 Container Security: Theory & Practice at Netflix
DCSF19 Container Security: Theory & Practice at NetflixDocker, Inc. Michael Wardrop discusses container security at Netflix, emphasizing the importance of tailored security solutions based on individual threat models. The document outlines the evolution of Netflix's container management platform, Titus, and highlights various security practices, including identity management, mutual TLS, and vulnerability management strategies. Key takeaways stress the significance of protecting cloud control planes, the necessity of involving security in the development lifecycle, and the importance of providing engineers with the tools to debug and monitor container performance.
Hands-on Helm
Hands-on Helm Docker, Inc. The document discusses Helm, a package manager for Kubernetes, providing an overview of its features and significant changes in the transition from Helm 2 to Helm 3. Key updates include improved security, simplified architecture, and modifications to CLI commands and release metadata management. The document also encourages community involvement and outlines considerations for migration and support of existing charts.
Using the SDACK Architecture on Security Event Inspection by Yu-Lun Chen and ...
Using the SDACK Architecture on Security Event Inspection by Yu-Lun Chen and ...Docker, Inc. The document presents insights on using the SDACK architecture for security event inspection, emphasizing the need for scalable, on-premises solutions to handle large data volumes and improve actionable insights from logs. It outlines the benefits of dockerizing applications for seamless deployment, testing, and scaling, along with considerations for security and configuration management. Key lessons learned highlight the importance of flexibility, reproducibility, and effective monitoring to enhance the development and operation of security-monitoring systems.
Puppet plugin for vRealize Automation (vRA)
Puppet plugin for vRealize Automation (vRA)Puppet The document discusses the integration of Puppet Enterprise and vRealize Automation (vRA) to enhance digital transformation, DevOps initiatives, and cloud migration. It highlights the benefits of automation, including increased deployment speed, reduced outages, and improved security and compliance measures. The collaboration aims to modernize IT processes, eliminate manual workflows, and provide a unified management platform for seamless operations.
Simple tweaks to get the most out of your JVM
Simple tweaks to get the most out of your JVMJamie Coleman The document discusses the Java Virtual Machine (JVM), its importance, architecture, and how cloud computing has influenced its use and optimization. It covers different JVM implementations including Hotspot, OpenJ9, and GraalVM, providing insights into their features and use cases. Additionally, it emphasizes performance tuning, efficient coding practices, and the balance of metrics for optimizing JVM performance in cloud environments.
Webinar: Accelerate Your Inner Dev Loop for Kubernetes Services
Webinar: Accelerate Your Inner Dev Loop for Kubernetes Services Ambassador Labs The document discusses an approach called Service Preview that allows developers to accelerate their inner development loops for Kubernetes services. Service Preview allows developers to code microservices locally, preview changes immediately with routed test traffic, use local tools like debuggers while treating the microservice as if it is live in the cluster. This is done in isolation before merging changes. Service Preview is presented as a simple way to remedy the complexities introduced to development workflows by containers and Kubernetes without the drawbacks of other approaches.
DCSF 19 How Entergy is Mitigating Legacy Windows Operating System Vulnerabili...
DCSF 19 How Entergy is Mitigating Legacy Windows Operating System Vulnerabili...Docker, Inc. The document outlines a large-scale IT security initiative by Entergy Corporation to mitigate vulnerabilities in legacy Windows operating systems using Docker Enterprise. It discusses the benefits of containerization, including reduced maintenance overhead and licensing costs, while integrating DevSecOps practices. Key lessons learned emphasize the need for transformation in large enterprises and the importance of project management and cultural change during this process.
DCSF 19 Microservices API: Routing Across Any Infrastructure
DCSF 19 Microservices API: Routing Across Any InfrastructureDocker, Inc. The document discusses the challenges of traditional infrastructure and the need for modern solutions such as microservices and API servers. It highlights issues with scaling, deployment methods, and routing across infrastructures while proposing solutions like orchestration with Kubernetes and Swarm. The focus is on achieving self-service redundancy, availability, and effective rate limiting in a production environment.
DCSF 19 Developing Apps with Containers, Functions and Cloud Services
DCSF 19 Developing Apps with Containers, Functions and Cloud ServicesDocker, Inc. Patrick Chanezon, a cloud advocate at Microsoft, discusses developing applications using containers, functions, and managed cloud services, emphasizing Microsoft's mission to empower organizations. He elaborates on cloud app abstractions including Docker for containers and Azure Functions for serverless computing, highlighting their features like scalability and event-driven capabilities. The document also covers development workflows, Kubernetes integration, and recent innovations in cloud-native application management like the Cloud Native Application Bundles (CNAB).
Serverless java
Serverless javaVishwas N 1. The document discusses serverless computing and its advantages over traditional server-based systems like scalability and reduced costs.
2. It notes some potential downsides of serverless like difficulties testing and debugging as well as security and vendor lock-in concerns.
3. The document provides an overview of serverless concepts like APIs, operations, scripting and functions and compares the serverless model to Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) models.
DCSF 19 Modernizing Insurance with Docker Enterprise: The Physicians Mutual ...
DCSF 19 Modernizing Insurance with Docker Enterprise: The Physicians Mutual ...Docker, Inc. The document outlines the modernization journey of Physicians Mutual's application development, detailing the transition from legacy systems to contemporary practices such as RESTful APIs and containerization using Docker. It emphasizes the importance of intentional collaboration and addressing stakeholder concerns to maintain momentum during the initiative. Outcomes include the establishment of numerous services and containers, improved high availability, and enhanced monitoring capabilities.
DCEU 18: From Monolith to Microservices
DCEU 18: From Monolith to MicroservicesDocker, Inc. The document discusses the challenges of transitioning from monolithic architectures to microservices, emphasizing release engineering, independent component lifecycles, and the impact of organizational structures on design. It outlines various tools and strategies for achieving operational excellence and managing complexity in microservices, including contract validation, dependency management, and continuous integration testing. Additionally, it highlights the significance of monitoring and health checks in distinguishing between single service and multiple service environments.
Accessible hpc for everyone with docker and containers
Accessible hpc for everyone with docker and containersDocker, Inc. The document discusses the advancements and challenges in high-performance computing (HPC), particularly with the integration of Docker and container technology. It highlights the evolution of HPC systems over decades and explains how containers can address barriers such as cost and expertise for scientists. Furthermore, it outlines current HPC scheduling methods, integration of AI, and opportunities for collaboration in the HPC community through various upcoming conferences.
DCEU 18: 5 Patterns for Success in Application Transformation
DCEU 18: 5 Patterns for Success in Application TransformationDocker, Inc. The document outlines five successful patterns for transforming Docker applications, emphasizing logging, configuration, dependencies, health checks, and metrics. It provides insights into how to enhance old applications without altering code or disrupting development workflows. Additionally, it includes links for further information on Docker trials and video resources for .NET and Java.
Networking in Docker EE 2.0 with Kubernetes and Swarm
Networking in Docker EE 2.0 with Kubernetes and SwarmAbhinandan P.b The document discusses Docker EE 2.0's networking capabilities with Kubernetes and Swarm, focusing on security, multi-tenancy, and observability. It highlights design principles, deployment options, and the flexibility of network configurations using various drivers. The conclusion emphasizes the choice between Swarm for simplicity and Kubernetes for greater flexibility and cloud integration.
Application Deployment and Management at Scale with 1&1 by Matt Baldwin
Application Deployment and Management at Scale with 1&1 by Matt BaldwinDocker, Inc. 1&1 manages application deployment and infrastructure for customers in a scalable, multi-tenant environment, addressing challenges of a monolithic architecture by moving towards a container-based model. The new architecture allows better isolation, independent scalability, and reduces management overhead while providing users with more customization options. Future goals include improved resource management, automated lifecycle management, and enhanced compatibility with various application components.
JEEconf 2017
JEEconf 2017Ihor Kolodyuk This document discusses automated scaling of JavaEE microservice stacks. It begins by introducing the speaker and their experience with autoscaling technologies. It then explains that while simple applications can be easily scaled by increasing container replicas, scaling complex JavaEE applications requires more work. The document outlines three key aspects needed for scalable applications: using lightweight application servers adapted for containers; decomposing monolithic applications into independently scalable microservices; and using orchestration software to automate provisioning, health checks, metrics collection, and scaling. It provides examples of scaling strategies for different application components like message queues and databases. It also discusses common issues like ensuring new instances are properly registered for events and services. The document emphasizes that proper configuration of autoscaling triggers
Container on azure
Container on azureVishwas N This document discusses containers and container orchestration on Azure. It begins with an introduction to containers and their advantages over virtual machines. It then covers building Dockerfiles, container commands, and hosting container registries and applications on Azure. Container orchestration with Kubernetes is discussed as a way to deploy and scale containerized applications on the cloud, providing capabilities like auto-scaling, self-healing, service discovery and load balancing. The document points to additional future content on using Azure Kubernetes Service.
Immutable Awesomeness by John Willis and Josh Corman
Immutable Awesomeness by John Willis and Josh CormanDocker, Inc. The document discusses the intersection of DevOps and immutable infrastructure, highlighting the importance of high performance organizational capital and the challenges in supply chain management. It analyzes various software vulnerabilities and approaches to ensure consistent and reliable infrastructure deployment, advocating for immutable delivery and principles like variety, velocity, variability, and visibility. Additionally, it references several materials and methodologies that can enhance agility and efficiency in DevOps practices.
Similar to Distributed Deep Learning with Docker at Salesforce (20)
No BS Guide to Deep Learning in the Enterprise
No BS Guide to Deep Learning in the EnterpriseJesus Rodriguez The document provides a comprehensive overview of deep learning in the enterprise, highlighting its challenges, methodologies, and current frameworks. It discusses various deep learning stack categories, including tiered frameworks and cloud solutions, emphasizing the complexity of implementation and the required expertise. The summary underscores that despite the crowded market, successful deep learning applications should start small and iterate quickly.
Why scala for data science
Why scala for data scienceGuglielmo Iozzia This document discusses why Scala is a good programming language for data science. It begins by providing background on Scala as a functional programming language that runs on the Java Virtual Machine. The main reasons given for using Scala in data science are its robustness for large datasets, integration with common big data tools that run on JVM, and available libraries like Spark MLlib, DeepLearning4J, and ND4J. Code examples are provided showing how to perform tasks with these libraries in Scala. The document also discusses how Scala, Python, and Keras can be used together via TensorFlow for prototyping models in Python and deploying them in Scala applications using DeepLearning4J.
Machine learning in the wild deployment
Machine learning in the wild deploymentBirger Moell This document provides an overview of machine learning concepts and deployment strategies. It discusses using Flask to host trained machine learning models, building an application with a trained Keras model, and testing the model using CURL requests. It also mentions Docker for deploying models as standalone applications and Kubernetes for automating container deployment and management. The document covers TensorFlow.js for loading pre-trained models in the browser and training models, as well as cloud platforms like AWS, Google, and Azure.
Canada DevOps Summit 2020 Presentation Nov_03_2020
Canada DevOps Summit 2020 Presentation Nov_03_2020Varun Manik Varun Kumar is a senior consultant at Deloitte with over 8 years of experience in cloud engineering and DevOps, holding several AWS certifications and a master's degree in computer science. He specializes in machine learning, cloud migrations, and automation, having recently led a cloud migration project for a media company in Southeast Asia. Varun is also an active technical blogger and trainer, contributing to the community through various platforms and training sessions.
Data Con LA 2018 - Towards Data Science Engineering Principles by Joerg Schad
Data Con LA 2018 - Towards Data Science Engineering Principles by Joerg SchadData Con LA This document discusses the need for data science engineering principles to standardize the machine learning development process. It outlines several challenges in machine learning projects including data quality issues, requirements engineering, continuous integration, testing, debugging, profiling, monitoring, and model serving. Standardizing these aspects into a repeatable pipeline can help address challenges like building reliable models, deploying models with zero downtime, and ensuring model quality during both training and serving. The document advocates for applying software engineering best practices to machine learning in order to more easily build and scale impactful AI technologies.
Notes on Deploying Machine-learning Models at Scale
Notes on Deploying Machine-learning Models at ScaleDeep Kayal The document discusses the complexities of deploying machine learning models at scale, emphasizing the importance of understanding hidden technical debt and the ongoing maintenance costs associated with real-world ML systems. It outlines a process for validating machine learning models through expert reviews and introduces a blueprint for deploying scalable APIs for both streaming and batch data processing. The author advocates for treating ML software as conventional software, highlighting the challenges and best practices for effective deployment and versioning.
Austin,TX Meetup presentation tensorflow final oct 26 2017
Austin,TX Meetup presentation tensorflow final oct 26 2017Clarisse Hedglin The document discusses a deep learning lecture by IBM, focusing on the TensorFlow framework and its growing popularity among data scientists due to its capabilities and extensive community support. It covers various frameworks, comparing their features and use cases, highlighting the significant impact of AI across multiple industries. Additionally, it addresses challenges in deep learning scalability and flexibility, emphasizing the need for effective distributed systems for optimal performance.
Deploying deep learning models with Docker and Kubernetes
Deploying deep learning models with Docker and KubernetesPetteriTeikariPhD This document discusses the deployment of deep learning models using Docker and Kubernetes, highlighting the challenges faced in both academic and startup settings regarding code reproducibility and legibility. It emphasizes the importance of microservices, containerization, and the role of Docker in simplifying the software development process within data science. Additionally, the document explores the shift towards Kubernetes for container orchestration and the benefits of adopting both platforms for efficient machine learning applications.
TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform Seldon 1. The document discusses building a data science platform on DC/OS to operationalize machine learning models. It outlines challenges at each stage of the ML pipeline and how DC/OS addresses them with distributed computing capabilities and services for data storage, processing, model training and deployment.
2. Key stages covered include data preparation, distributed training using frameworks like TensorFlow, model management with storage of trained models, and low-latency model serving for production with TensorFlow Serving.
3. DC/OS provides a full-stack platform to operationalize ML at scale through distributed computing resources, container orchestration, and integration of open source data and ML services.
End-to-End Platform Support for Distributed Deep Learning in Finance
End-to-End Platform Support for Distributed Deep Learning in FinanceJim Dowling Distributed deep learning platforms can improve productivity for practitioners by enabling faster training through distributed computing. Hopsworks is a new data platform built on HopsFS that provides first-class support for Python, deep learning, machine learning, and strong data governance. It allows for distributed deep learning training and parallel experiments across GPU servers as well as secure streaming analytics using Kafka, Spark, and Flink.
BigDL Deep Learning in Apache Spark - AWS re:invent 2017
BigDL Deep Learning in Apache Spark - AWS re:invent 2017Dave Nielsen The document outlines large-scale deep learning using BigDL, a distributed deep learning library for Apache Spark. It discusses the evolution and applications of artificial intelligence, machine learning, and deep learning, emphasizing modern frameworks and tools like TensorFlow and Amazon Machine Learning. BigDL enables scalable deep learning operations while integrating with big data ecosystems, offering high performance and cost-effective solutions for processing large datasets.
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019Provectus The document provides a comprehensive guide on scaling machine learning (ML) from initial development to deployment for millions of users using AWS services. It highlights the importance of automation, offers various deployment strategies including EC2, ECS, EKS, and SageMaker, and discusses the trade-offs associated with each option. Recommendations are made on using AWS tools to optimize ML workflows while emphasizing the need to focus on models and data over infrastructure complications.
DevOps for Machine Learning overview en-us
DevOps for Machine Learning overview en-useltonrodriguez11 This document discusses applying DevOps practices and principles to machine learning model development and deployment. It outlines how continuous integration (CI), continuous delivery (CD), and continuous monitoring can be used to safely deliver ML features to customers. The benefits of this approach include continuous value delivery, end-to-end ownership by data science teams, consistent processes, quality/cadence improvements, and regulatory compliance. Key aspects covered are experiment tracking, model versioning, packaging and deployment, and monitoring models in production.
Scalable and Distributed DNN Training on Modern HPC Systems
Scalable and Distributed DNN Training on Modern HPC Systemsinside-BigData.com This document discusses scaling deep learning training on HPC systems. It begins by providing background on deep learning and how interest in it has grown significantly. It then discusses how HPC systems can be leveraged for deep learning by supporting distributed training across multiple nodes. Several challenges of designing deep learning frameworks for HPC are outlined, including memory and communication overhead. The document proposes a co-design approach between deep learning frameworks and communication runtimes to better support distributed training and exploit HPC resources. MVAPICH2 software is discussed as an example that provides optimized MPI support for CPU- and GPU-based deep learning on HPC clusters.
IBM Developer Model Asset eXchange
IBM Developer Model Asset eXchangeNick Pentreath The document discusses the IBM Developer Model Asset Exchange and the Fabric for Deep Learning (FFDL), emphasizing its role in simplifying the creation and management of AI solutions in enterprises. FFDL provides a consistent, scalable, and fault-tolerant framework for training and deploying deep learning models across various frameworks. It also highlights the availability of free, open-source deep learning models and the integration of multiple domains and technologies for AI development.
Scaling Deep Learning Algorithms on Extreme Scale Architectures
Scaling Deep Learning Algorithms on Extreme Scale Architecturesinside-BigData.com This document summarizes a presentation on scaling deep learning algorithms on extreme scale architectures. It discusses challenges in using deep learning, a vision for machine/deep learning R&D including novel algorithms, and the MaTEx toolkit which supports distributed deep learning on GPU and CPU clusters. Sample results show strong and weak scaling of asynchronous gradient descent on Summit. Fault tolerance needs and the impact of deep learning on other domains are also covered.
Deep learning with kafka
Deep learning with kafkaNitin Kumar 1. The document discusses using Deeplearning4j and Kafka together for machine learning workflows. It describes how Deeplearning4j can be used to build, train, and deploy neural networks on JVM and Spark, while Kafka can be used to stream data for training and inference.
2. An example application is described that performs anomaly detection on log files from a CDN by aggregating the data to reduce the number of data points. This allows the model to run efficiently on available GPU hardware.
3. The document provides a link to a GitHub repository with a code example that uses Kafka to stream data, Keras to train a model, and Deeplearning4j to perform inference in Java and deploy the model.
Kubeflow.pptx
Kubeflow.pptxdhaferbenali1 The document discusses challenges with enterprise machine learning and proposes Kubeflow as a solution. It notes that data scientists prefer different tools, models are difficult to deploy and manage at scale, and a shared platform is needed. Kubeflow is presented as providing a "lab and factory" environment to explore ideas and reproducibly run models at scale using containers, Kubernetes, notebooks, pipelines and other tools. It aims to help machine learning models progress from research to production.
Data ops: Machine Learning in production
Data ops: Machine Learning in productionStepan Pushkarev The document discusses the challenges and solutions for deploying machine learning in production, focusing on the architecture, data quality, and monitoring aspects. It highlights the importance of a unified framework for data processing and experimentation to ensure reliability and efficiency in model deployment. Key takeaways include the need for schema-first design, automated data profiling, and predictive retraining to minimize errors and enhance machine learning operations.
Explore Deep Learning Architecture using Tensorflow 2.0 now! Part 2
Explore Deep Learning Architecture using Tensorflow 2.0 now! Part 2Tyrone Systems The document outlines a live webinar on deep learning architecture using TensorFlow scheduled for May 6, 2020. Key topics include the use of GPU systems optimized for deep learning, building convolutional and sequence models, exploring generative models, and model quantization for edge devices. The document also details the hardware and software products offered by Tyrone Systems that support deep learning deployments.
Ad
More from Docker, Inc. (20)
Containerize Your Game Server for the Best Multiplayer Experience
Containerize Your Game Server for the Best Multiplayer Experience Docker, Inc. The document discusses the benefits of containerizing game servers to enhance multiplayer gaming experiences, emphasizing the importance of low latency and performance. It outlines challenges such as managing global deployments and the need for persistent data, while highlighting the advantages of using Docker containers for deployment and management. Additionally, Agones is introduced as a solution for orchestrating game servers within a Kubernetes environment.
How to Improve Your Image Builds Using Advance Docker Build
How to Improve Your Image Builds Using Advance Docker BuildDocker, Inc. The document presents an overview of BuildKit, a build engine for Docker, highlighting its features such as multi-stage builds, improved build caching, build secrets management, and SSH agent forwarding. It details the advantages of using BuildKit over legacy build engines, including performance improvements and the ability to manage dependencies more efficiently. The document also includes demo instructions and practical use cases for transitioning to BuildKit in various development scenarios.
Build & Deploy Multi-Container Applications to AWS
Build & Deploy Multi-Container Applications to AWSDocker, Inc. The document outlines the process of building and deploying multi-container applications to AWS using Docker Compose. It includes details about containerizing various services like an Nginx server, a React app, and a Node.js server, as well as configuration steps for AWS deployment. The document emphasizes the use of a CLI tool to simplify the management of multiple containers.
Securing Your Containerized Applications with NGINX
Securing Your Containerized Applications with NGINXDocker, Inc. The document summarizes Kevin Jones' presentation on securing containerized applications with NGINX. It discusses the benefits of using a reverse proxy for security, NGINX best practices for TLS configuration, and deploying NGINX in Docker containers. It also provides code examples and configurations for setting up NGINX as a reverse proxy, optimizing TLS, and using NGINX as a sidecar proxy.
How To Build and Run Node Apps with Docker and Compose
How To Build and Run Node Apps with Docker and ComposeDocker, Inc. The document provides a comprehensive guide on building and running Node.js applications using Docker and Docker Compose. It covers essential steps such as creating Dockerfiles, setting up service definitions, and managing environment variables, along with deployment considerations. It also highlights best practices for containerization, including the use of volumes and securing applications with Let's Encrypt.
The First 10M Pulls: Building The Official Curl Image for Docker Hub
The First 10M Pulls: Building The Official Curl Image for Docker HubDocker, Inc. The document discusses the creation and implementation of the official curl image for Docker Hub, detailing the design decisions, security considerations, and the build process. It highlights the use of multi-stage builds for minimal image size and includes references to security vulnerabilities and testing procedures. Additionally, the document outlines lessons learned and future steps for enhancing the curl docker image.
COVID-19 in Italy: How Docker is Helping the Biggest Italian IT Company Conti...
COVID-19 in Italy: How Docker is Helping the Biggest Italian IT Company Conti...Docker, Inc. The document discusses how Engineering Ingegneria Informatica leveraged Docker technology to maintain operational continuity during the COVID-19 pandemic, managing 250,000 remote workplaces for 400 clients. It highlights their digital workplace infrastructure, DevOps culture, and educational offerings revolving around Docker training and certifications. Additionally, it outlines their biosurveillance solution, Eng-De4Bios, designed to monitor and geolocate COVID-19 cases and prevent the spread of the virus.
Become a Docker Power User With Microsoft Visual Studio Code
Become a Docker Power User With Microsoft Visual Studio CodeDocker, Inc. The document is an agenda for a presentation by Brian Christner on becoming a Docker power user using Microsoft Visual Studio Code. It covers topics such as integrating Docker Desktop with VS Code, utilizing various features of the Docker extension, and includes demos and resources for further learning. Key highlights include bootstrapping new Docker projects, writing Dockerfiles with code completion, and the ability to debug Docker applications.
How to Use Mirroring and Caching to Optimize your Container Registry
How to Use Mirroring and Caching to Optimize your Container RegistryDocker, Inc. The document discusses the use of caching and mirroring to optimize image registries, highlighting methods like configuring registry mirrors and manual syncing of images. It addresses limitations and risks associated with caching, emphasizing the benefits of managing your own mirror to save bandwidth and ensure resilience against upstream outages. Additionally, it provides technical implementation details and strategies for manually mirroring Docker images.
Kubernetes at Datadog Scale
Kubernetes at Datadog ScaleDocker, Inc. Kubernetes networking can be complex to scale due to issues like growing iptables rules, but newer solutions are helping. Pod networking uses CNI plugins like flannel or Calico to assign each pod an IP and allow communication. Service networking uses kube-proxy and iptables or IPVS for load balancing to pods. DNS is used to resolve service names to IPs. While Kubernetes networking brings flexibility, operators must learn the nuances of their specific CNI plugin and issues can arise, but the ecosystem adapts quickly to new needs and changes don't impact all workloads.
Labels, Labels, Labels
Labels, Labels, Labels Docker, Inc. The document discusses the importance of using labels in software development for security, simplicity, and documentation. It provides specific examples of label schemas for container images, including details about authors, version control, and build information. Additionally, it outlines how to create and view these labels using Docker and Kubernetes commands.
Using Docker Hub at Scale to Support Micro Focus' Delivery and Deployment Model
Using Docker Hub at Scale to Support Micro Focus' Delivery and Deployment ModelDocker, Inc. Micro Focus uses Docker Hub at scale to support its software delivery and deployment model. Some key points:
- Docker Hub is used as the registry service for Micro Focus container images
- It allows for optimized, secure, reliable and cost-effective software delivery through deployments and updates of container images to customers and partners
- Micro Focus leverages features like private repositories, offline/online access, signing and scanning of images, and integration with CI/CD pipelines
- Over 1,650 organizations, 450 repositories, and 18 teams are used on Docker Hub to manage access and deliver software from Micro Focus
Build & Deploy Multi-Container Applications to AWS
Build & Deploy Multi-Container Applications to AWSDocker, Inc. The document outlines the process of building and deploying multi-container applications to AWS using Docker Compose. It details the steps involved, including containerizing an NGINX server, a React app, and a Node.js server, as well as configuring files such as buildspec.yml and dockerrun.aws.json. The summary also mentions the importance of environment variables in AWS Elastic Beanstalk and triggering the deployment pipeline.
From Fortran on the Desktop to Kubernetes in the Cloud: A Windows Migration S...
From Fortran on the Desktop to Kubernetes in the Cloud: A Windows Migration S...Docker, Inc. The document discusses Elton Stoneman's experiences migrating from Fortran on desktop to Kubernetes in the cloud, focusing on themes such as flexibility, speed, and standardization in app modeling. It outlines a structured migration process with phases for proof of concept and evaluation, involving the development of compute modules and APIs. Additionally, it includes promotional links for learning resources on Docker and Kubernetes.
Developing with Docker for the Arm Architecture
Developing with Docker for the Arm ArchitectureDocker, Inc. This document outlines the advantages of using Docker for ARM architecture development, emphasizing its ability to simplify and modernize the development process. It discusses the integration of Docker with various tools and platforms, including AWS and Raspberry Pi, and highlights the benefits of multi-architecture support and template automation. Additionally, it presents practical exercises for developers to enhance their skills in building and deploying applications using Docker.
Sharing is Caring: How to Begin Speaking at Conferences
Sharing is Caring: How to Begin Speaking at ConferencesDocker, Inc. The document outlines strategies for speaking at conferences, emphasizing the importance of sharing knowledge for career growth and networking. It provides guidance on preparing proposals, creating engaging content, and delivering effective presentations, including tips for overcoming fear and soliciting feedback. The session aims to equip participants with the necessary skills and confidence to present successfully at technical events.
Virtual Meetup Docker + Arm: Building Multi-arch Apps with Buildx
Virtual Meetup Docker + Arm: Building Multi-arch Apps with BuildxDocker, Inc. This document discusses building multi-architecture Docker images using Docker Buildx. It introduces the concepts of multi-architecture images, describes the traditional method of building separate images for each architecture and combining them with manifest lists, and recommends using Buildx which can build images for multiple architectures simultaneously during a single build. Buildx allows building on the host architecture natively or emulating other architectures. The document emphasizes that applications should support multiple platforms and architectures, and encourages always using multi-stage Dockerfiles and considering Buildx as an improved alternative to the older manifest-based approach.
DCSF 19 eBPF Superpowers
DCSF 19 eBPF SuperpowersDocker, Inc. The document discusses eBPF (extended Berkeley Packet Filter) as a powerful tool for running code in the Linux kernel without needing to write traditional kernel modules, allowing flexibility and performance improvements. It covers the structure of eBPF programs, how they can interact with kernel and user-space, the various ways to attach these programs to events, and the role of helper functions and maps in data handling. Additionally, it mentions limitations of eBPF such as restrictions on loops and certain data types to ensure safe execution.
DCSF 19 Zero Trust Networks Come to Enterprise Kubernetes
DCSF 19 Zero Trust Networks Come to Enterprise KubernetesDocker, Inc. The document discusses a presentation about implementing zero trust networks on Docker Enterprise Kubernetes. It begins with motivations for zero trust like changes in app architectures, security threats, and deficiencies of traditional network zoning models. It then covers using Calico and Istio on Docker Enterprise to provide zero trust security with benefits like resilience against compromise and decoupling security from network location. The presentation includes a demo of a sample app and concludes with time for questions.
DCSF 19 Node.js Rocks in Docker for Dev and Ops
DCSF 19 Node.js Rocks in Docker for Dev and OpsDocker, Inc. The document outlines best practices for using Node.js with Docker, including guidelines for writing Dockerfiles, managing Node.js processes, and implementing multi-stage builds. It emphasizes proper shutdown procedures, security considerations, and the use of Docker Compose for managing multi-service applications. Additionally, it addresses common issues like handling node_modules in Docker and ensures a solid production checklist for Node.js applications.
Ad
Recently uploaded (20)
FIDO Seminar: Authentication for a Billion Consumers - Amazon.pptx
FIDO Seminar: Authentication for a Billion Consumers - Amazon.pptxFIDO Alliance FIDO Seminar: Authentication for a Billion Consumers - Amazon
National Fuels Treatments Initiative: Building a Seamless Map of Hazardous Fu...
National Fuels Treatments Initiative: Building a Seamless Map of Hazardous Fu...Safe Software The National Fuels Treatments Initiative (NFT) is transforming wildfire mitigation by creating a standardized map of nationwide fuels treatment locations across all land ownerships in the United States. While existing state and federal systems capture this data in diverse formats, NFT bridges these gaps, delivering the first truly integrated national view. This dataset will be used to measure the implementation of the National Cohesive Wildland Strategy and demonstrate the positive impact of collective investments in hazardous fuels reduction nationwide. In Phase 1, we developed an ETL pipeline template in FME Form, leveraging a schema-agnostic workflow with dynamic feature handling intended for fast roll-out and light maintenance. This was key as the initiative scaled from a few to over fifty contributors nationwide. By directly pulling from agency data stores, oftentimes ArcGIS Feature Services, NFT preserves existing structures, minimizing preparation needs. External mapping tables ensure consistent attribute and domain alignment, while robust change detection processes keep data current and actionable. Now in Phase 2, we’re migrating pipelines to FME Flow to take advantage of advanced scheduling, monitoring dashboards, and automated notifications to streamline operations. Join us to explore how this initiative exemplifies the power of technology, blending FME, ArcGIS Online, and AWS to solve a national business problem with a scalable, automated solution.
Supporting the NextGen 911 Digital Transformation with FME
Supporting the NextGen 911 Digital Transformation with FMESafe Software Next Generation 911 involves the transformation of our 911 system from an old analog one to the new digital internet based architecture. The evolution of NG911 opens up a host of new opportunities to improve the system. This includes everything from device based location, to real time text. This can improve location accuracy dramatically as well as provide live updates from the citizen in need along with real time sensor updates. There is also the opportunity to provide multi-media attachments and medical records if the end user approves. This digital transformation and enhancements all require the support of new NENA and CRTC standards, along with integration across a variety of data streams.
This presentation will focus on how FME has supported NG911 transformations to date, and how we are positioning FME to support the enhanced capabilities to come. This session will be of interest to emergency services, municipalities and anyone who may be interested to know more about how emergency services are being improved to provide more accurate, localized information in order to improve the speed and relevance of emergency response and ultimately save more lives and provide better outcomes for those in need.
Tech-ASan: Two-stage check for Address Sanitizer - Yixuan Cao.pdf
Tech-ASan: Two-stage check for Address Sanitizer - Yixuan Cao.pdfcaoyixuan2019 A presentation at Internetware 2025.
Floods in Valencia: Two FME-Powered Stories of Data Resilience
Floods in Valencia: Two FME-Powered Stories of Data ResilienceSafe Software In October 2024, the Spanish region of Valencia faced severe flooding that underscored the critical need for accessible and actionable data. This presentation will explore two innovative use cases where FME facilitated data integration and availability during the crisis. The first case demonstrates how FME was used to process and convert satellite imagery and other geospatial data into formats tailored for rapid analysis by emergency teams. The second case delves into making human mobility data—collected from mobile phone signals—accessible as source-destination matrices, offering key insights into population movements during and after the flooding. These stories highlight how FME's powerful capabilities can bridge the gap between raw data and decision-making, fostering resilience and preparedness in the face of natural disasters. Attendees will gain practical insights into how FME can support crisis management and urban planning in a changing climate.
“Why It’s Critical to Have an Integrated Development Methodology for Edge AI,...
“Why It’s Critical to Have an Integrated Development Methodology for Edge AI,...Edge AI and Vision Alliance For the full video of this presentation, please visit: https://www.edge-ai-vision.com/2025/06/why-its-critical-to-have-an-integrated-development-methodology-for-edge-ai-a-presentation-from-lattice-semiconductor/
Sreepada Hegade, Director of ML Systems and Software at Lattice Semiconductor, presents the “Why It’s Critical to Have an Integrated Development Methodology for Edge AI” tutorial at the May 2025 Embedded Vision Summit.
The deployment of neural networks near sensors brings well-known advantages such as lower latency, privacy and reduced overall system cost—but also brings significant challenges that complicate development. These challenges can be addressed effectively by choosing the right solution and design methodology. The low-power FPGAs from Lattice are well poised to enable efficient edge implementation of models, while Lattice’s proven development methodology helps to mitigate the challenges and risks associated with edge model deployment.
In this presentation, Hegade explains the importance of an integrated framework that tightly consolidates different aspects of edge AI development, including training, quantization of networks for edge deployment, integration with sensors and inferencing. He also illustrates how Lattice’s simplified tool flow helps to achieve the best trade-off between power, performance and efficiency using low-power FPGAs for edge deployment of various AI workloads.
High Availability On-Premises FME Flow.pdf
High Availability On-Premises FME Flow.pdfSafe Software FME Flow is a highly robust tool for transforming data both automatically and by user-initiated workflows. At the Finnish telecommunications company Elisa, FME Flow serves processes and internal stakeholders that require 24/7 availability from underlying systems, while imposing limitations on the use of cloud based systems. In response to these business requirements, Elisa has implemented a high-availability on-premises setup of FME Flow, where all components of the system have been duplicated or clustered. The goal of the presentation is to provide insights into the architecture behind the high-availability functionality. The presentation will show in basic technical terms how the different parts of the system work together. Basic level understanding of IT technologies is required to understand the technical portion of the presentation, namely understanding the purpose of the following components: load balancer, FME Flow host nodes, FME Flow worker nodes, network file storage drives, databases, and external authentication services. The presentation will also outline our lessons learned from the high-availability project, both benefits and challenges to consider.
Crypto Super 500 - 14th Report - June2025.pdf
Crypto Super 500 - 14th Report - June2025.pdfStephen Perrenod This OrionX's 14th semi-annual report on the state of the cryptocurrency mining market. The report focuses on Proof-of-Work cryptocurrencies since those use substantial supercomputer power to mint new coins and encode transactions on their blockchains. Only two make the cut this time, Bitcoin with $18 billion of annual economic value produced and Dogecoin with $1 billion. Bitcoin has now reached the Zettascale with typical hash rates of 0.9 Zettahashes per second. Bitcoin is powered by the world's largest decentralized supercomputer in a continuous winner take all lottery incentive network.
TrustArc Webinar - 2025 Global Privacy Survey
TrustArc Webinar - 2025 Global Privacy SurveyTrustArc How does your privacy program compare to your peers? What challenges are privacy teams tackling and prioritizing in 2025?
In the sixth annual Global Privacy Benchmarks Survey, we asked global privacy professionals and business executives to share their perspectives on privacy inside and outside their organizations. The annual report provides a 360-degree view of various industries' priorities, attitudes, and trends. See how organizational priorities and strategic approaches to data security and privacy are evolving around the globe.
This webinar features an expert panel discussion and data-driven insights to help you navigate the shifting privacy landscape. Whether you are a privacy officer, legal professional, compliance specialist, or security expert, this session will provide actionable takeaways to strengthen your privacy strategy.
This webinar will review:
- The emerging trends in data protection, compliance, and risk
- The top challenges for privacy leaders, practitioners, and organizations in 2025
- The impact of evolving regulations and the crossroads with new technology, like AI
Predictions for the future of privacy in 2025 and beyond
Kubernetes Security Act Now Before It’s Too Late
Kubernetes Security Act Now Before It’s Too LateMichael Furman In today's cloud-native landscape, Kubernetes has become the de facto standard for orchestrating containerized applications, but its inherent complexity introduces unique security challenges. Are you one YAML away from disaster?
This presentation, "Kubernetes Security: Act Now Before It’s Too Late," is your essential guide to understanding and mitigating the critical security risks within your Kubernetes environments. This presentation dives deep into the OWASP Kubernetes Top Ten, providing actionable insights to harden your clusters.
We will cover:
The fundamental architecture of Kubernetes and why its security is paramount.
In-depth strategies for protecting your Kubernetes Control Plane, including kube-apiserver and etcd.
Crucial best practices for securing your workloads and nodes, covering topics like privileged containers, root filesystem security, and the essential role of Pod Security Admission.
Don't wait for a breach. Learn how to identify, prevent, and respond to Kubernetes security threats effectively.
It's time to act now before it's too late!
“Addressing Evolving AI Model Challenges Through Memory and Storage,” a Prese...
“Addressing Evolving AI Model Challenges Through Memory and Storage,” a Prese...Edge AI and Vision Alliance For the full video of this presentation, please visit: https://www.edge-ai-vision.com/2025/06/addressing-evolving-ai-model-challenges-through-memory-and-storage-a-presentation-from-micron/
Wil Florentino, Senior Segment Marketing Manager at Micron, presents the “Addressing Evolving AI Model Challenges Through Memory and Storage” tutorial at the May 2025 Embedded Vision Summit.
In the fast-changing world of artificial intelligence, the industry is deploying more AI compute at the edge. But the growing diversity and data footprint of transformers and models such as large language models and large multimodal models puts a spotlight on memory performance and data storage capacity as key bottlenecks. Enabling the full potential of AI in industries such as manufacturing, automotive, robotics and transportation will require us to find efficient ways to deploy this new generation of complex models.
In this presentation, Florentino explores how memory and storage are responding to this need and solving complex issues in the AI market. He examines the storage capacity and memory bandwidth requirements of edge AI use cases ranging from tiny devices with severe cost and power constraints to edge servers, and he explains how new memory technologies such as LPDDR5, LPCAMM2 and multi-port SSDs are helping system developers to meet these challenges.
Can We Use Rust to Develop Extensions for PostgreSQL? (POSETTE: An Event for ...
Can We Use Rust to Develop Extensions for PostgreSQL? (POSETTE: An Event for ...NTT DATA Technology & Innovation Can We Use Rust to Develop Extensions for PostgreSQL?
(POSETTE: An Event for Postgres 2025)
June 11, 2025
Shinya Kato
NTT DATA Japan Corporation
ENERGY CONSUMPTION CALCULATION IN ENERGY-EFFICIENT AIR CONDITIONER.pdf
ENERGY CONSUMPTION CALCULATION IN ENERGY-EFFICIENT AIR CONDITIONER.pdfMuhammad Rizwan Akram DC Inverter Air Conditioners are revolutionizing the cooling industry by delivering affordable,
energy-efficient, and environmentally sustainable climate control solutions. Unlike conventional
fixed-speed air conditioners, DC inverter systems operate with variable-speed compressors that
modulate cooling output based on demand, significantly reducing energy consumption and
extending the lifespan of the appliance.
These systems are critical in reducing electricity usage, lowering greenhouse gas emissions, and
promoting eco-friendly technologies in residential and commercial sectors. With advancements in
compressor control, refrigerant efficiency, and smart energy management, DC inverter air conditioners
have become a benchmark in sustainable climate control solutions
“Key Requirements to Successfully Implement Generative AI in Edge Devices—Opt...
“Key Requirements to Successfully Implement Generative AI in Edge Devices—Opt...Edge AI and Vision Alliance For the full video of this presentation, please visit: https://www.edge-ai-vision.com/2025/06/key-requirements-to-successfully-implement-generative-ai-in-edge-devices-optimized-mapping-to-the-enhanced-npx6-neural-processing-unit-ip-a-presentation-from-synopsys/
Gordon Cooper, Principal Product Manager at Synopsys, presents the “Key Requirements to Successfully Implement Generative AI in Edge Devices—Optimized Mapping to the Enhanced NPX6 Neural Processing Unit IP” tutorial at the May 2025 Embedded Vision Summit.
In this talk, Cooper discusses emerging trends in generative AI for edge devices and the key role of transformer-based neural networks. He reviews the distinct attributes of transformers, their advantages over conventional convolutional neural networks and how they enable generative AI.
Cooper then covers key requirements that must be met for neural processing units (NPU) to support transformers and generative AI in edge device applications. He uses transformer-based generative AI examples to illustrate the efficient mapping of these workloads onto the enhanced Synopsys ARC NPX NPU IP family.
Integration of Utility Data into 3D BIM Models Using a 3D Solids Modeling Wor...
Integration of Utility Data into 3D BIM Models Using a 3D Solids Modeling Wor...Safe Software Jacobs has developed a 3D utility solids modelling workflow to improve the integration of utility data into 3D Building Information Modeling (BIM) environments. This workflow, a collaborative effort between the New Zealand Geospatial Team and the Australian Data Capture Team, employs FME to convert 2D utility data into detailed 3D representations, supporting enhanced spatial analysis and clash detection.
To enable the automation of this process, Jacobs has also developed a survey data standard that standardizes the capture of existing utilities. This standard ensures consistency in data collection, forming the foundation for the subsequent automated validation and modelling steps. The workflow begins with the acquisition of utility survey data, including attributes such as location, depth, diameter, and material of utility assets like pipes and manholes. This data is validated through a custom-built tool that ensures completeness and logical consistency, including checks for proper connectivity between network components. Following validation, the data is processed using an automated modelling tool to generate 3D solids from 2D geometric representations. These solids are then integrated into BIM models to facilitate compatibility with 3D workflows and enable detailed spatial analyses.
The workflow contributes to improved spatial understanding by visualizing the relationships between utilities and other infrastructure elements. The automation of validation and modeling processes ensures consistent and accurate outputs, minimizing errors and increasing workflow efficiency.
This methodology highlights the application of FME in addressing challenges associated with geospatial data transformation and demonstrates its utility in enhancing data integration within BIM frameworks. By enabling accurate 3D representation of utility networks, the workflow supports improved design collaboration and decision-making in complex infrastructure projects
The State of Web3 Industry- Industry Report
The State of Web3 Industry- Industry ReportLiveplex Web3 is poised for mainstream integration by 2030, with decentralized applications potentially reaching billions of users through improved scalability, user-friendly wallets, and regulatory clarity. Many forecasts project trillions of dollars in tokenized assets by 2030 , integration of AI, IoT, and Web3 (e.g. autonomous agents and decentralized physical infrastructure), and the possible emergence of global interoperability standards. Key challenges going forward include ensuring security at scale, preserving decentralization principles under regulatory oversight, and demonstrating tangible consumer value to sustain adoption beyond speculative cycles.
AI vs Human Writing: Can You Tell the Difference?
AI vs Human Writing: Can You Tell the Difference?Shashi Sathyanarayana, Ph.D This slide illustrates a side-by-side comparison between human-written, AI-written, and ambiguous content. It highlights subtle cues that help readers assess authenticity, raising essential questions about the future of communication, trust, and thought leadership in the age of generative AI.
FIDO Seminar: Targeting Trust: The Future of Identity in the Workforce.pptx
FIDO Seminar: Targeting Trust: The Future of Identity in the Workforce.pptxFIDO Alliance FIDO Seminar: Targeting Trust: The Future of Identity in the Workforce
FIDO Seminar: New Data: Passkey Adoption in the Workforce.pptx
FIDO Seminar: New Data: Passkey Adoption in the Workforce.pptxFIDO Alliance FIDO Seminar: New Data: Passkey Adoption in the Workforce
“Why It’s Critical to Have an Integrated Development Methodology for Edge AI,...
“Why It’s Critical to Have an Integrated Development Methodology for Edge AI,...Edge AI and Vision Alliance
“Addressing Evolving AI Model Challenges Through Memory and Storage,” a Prese...
“Addressing Evolving AI Model Challenges Through Memory and Storage,” a Prese...Edge AI and Vision Alliance
Can We Use Rust to Develop Extensions for PostgreSQL? (POSETTE: An Event for ...
Can We Use Rust to Develop Extensions for PostgreSQL? (POSETTE: An Event for ...NTT DATA Technology & Innovation
“Key Requirements to Successfully Implement Generative AI in Edge Devices—Opt...
“Key Requirements to Successfully Implement Generative AI in Edge Devices—Opt...Edge AI and Vision Alliance
Distributed Deep Learning with Docker at Salesforce
- 1. Distributed Deep Learning with Docker at Salesforce
- 3. Caveats ● These my own views and opinions, not those of Salesforce ● This is how one team at Salesforce deploys deep learning models ● When I use the term Docker I am referring to the technology, not the company ● Some of these designs are simplified
- 4. ● What is deep learning and why is it difficult? ● Deep learning at Salesforce ● Challenges ○ Designing for team specialization ○ Interacting with the model server ○ Testing ● Key takeaways About this talk
- 5. The core task of deep learning is function approximation. Neural networks can approximate any function. Neural networks are expensive to evaluate. ● Linear regression: ~1,000 parameters ● Deep neural network: 100M - 1B parameters (100,000 - 1M x linear reg.) Deep Learning Review
- 6. How should we design distributed systems for deep learning? high latency tasks
- 7. We use deep learning models to provide our customers useful information about their sales process. They send us this data as a firehose of streaming data. The faster we get this data to our customers, the more useful and actionable it is for their sales teams. Deep Learning at Salesforce
- 8. There are three steps to this process 1. Preprocessing - cleaning and formatting the data 2. Inference - running the data through the model 3. Postprocessing - interpreting the output from the model Deep Learning at Salesforce
- 9. Deep Learning at Salesforce Discusses cat preprocess [0.2, 0.71, 0.89, 0.6] [0.85, 0.15] inference postprocess“Hello! My cat is friendly.”
- 10. Deep Learning at Salesforce Discusses cat preprocess [0.2, 0.71, 0.89, 0.6] [0.85, 0.15] inference postprocess“Hello! My cat is friendly.” Data Science Team
- 11. Deep Learning at Salesforce Discusses cat preprocess [0.2, 0.71, 0.89, 0.6] [0.85, 0.15] inference postprocess“Hello! My cat is friendly.” Data Science Team Systems Team
- 13. Requirements 1. The data science team shouldn’t need to know about the system. They just want to define a sequence of computation. 2. The systems engineers shouldn’t need to know anything about the computation. They just want to scale the system. Designing for team specialization
- 14. Challenges 1. Some functions takes longer to execute than others (e.g., model inference) 2. The order of execution is important Designing for team specialization
- 15. Solution: map functions to containers Designing for team specialization postprocess(inference(preprocess(x))) preprocess inference postprocess
- 16. What about throughput? preprocess inference postprocess 0110010 1001111 0110010 1000011 It’s a cat! 1,000 QPS 500 QPS 300 QPS 1,000 QPS Max throughput
- 17. inference inferenceinferencepreprocess What about throughput? preprocess inference postprocess 0110010 1001111 0110010 1000011 It’s a cat! 1,000 QPS 2x 500 QPS 4x 300 QPS 1,000 QPS Max throughput Docker enables us to easily scale out each individual stage
- 18. inferenceinferencepreprocess What about throughput? preprocess inference postprocess 0110010 1001111 0110010 1000011 It’s a cat! 1,000 QPS 2x 500 QPS 2x 300 QPS 1,000 QPS Max throughput Kafka gives stage-wise checkpointing
- 19. Challenge 2: interacting with the model servers
- 20. Model servers provide a way to query the model, typically via gRPC or HTTP. What is the best way to deploy and interact with these model servers? Serving deep learning models
- 21. Challenge: 1. Model servers are designed as a standalone process. 2. How should we best utilize multiple GPUs? 3. What about networking? Interacting with the model server We want to keep deployment simple!
- 22. Solution: Deploy model server images as part of a “pod” or “group” with a coordinator service Interacting with the model server JVM Manager Model Server Model Server Model Server... Pod
- 23. 1. Who owns the model server? 2. How should we handle model versions? Where are they stored locally? 3. What are the addresses of the model servers? This solves additional challenges Data science team Docker shared volume http://localhost via Docker private networking
- 25. Challenge: how should we test these systems? 1. Deep learning models are probabilistic 2. Interservice interactions can be quite complex Testing
- 26. Solution: Docker Compose ● Makes it easy to swap out the model server with a mock service ● Deploying the entire system locally is easy ● Integrates well with Maven and Gradle Testing
- 27. We haven’t spent a lot of time discussing the details of Docker That is precisely the point!
- 28. ● Docker allows us to simplify many aspects of our design. ● Docker stays out of the way. ● Docker provides a simple alternative to a much more complex solution. Docker simplifies our lives