This document provides a summary of lessons learned from deploying Apache Spark as a cloud service on IBM Power Systems. Key points include:
1. Using an open source stack enabled agile development. Continuous integration automated deployment.
2. Platform Symphony efficiently allocated resources for multi-tenancy.
3. Performance tests found Spark ran 1.7x faster on Power than x86, improving economics.
4. Potential for acceleration using Power features like CAPI flash was identified.
Putting these lessons together, Power Systems could differentiate cloud data services through improved cost performance, agile development, and advanced acceleration.
Beyond Ingresses - Better Traffic Management in KubernetesMark McBride
Kubernetes makes deploying code easy, but conflating deploys and releases is risky. Using smarter proxies you can dramatically reduce the risk of a release, which in turn helps you ship code to customers faster.
Docker storage designing a platform for persistent dataDocker, Inc.
Docker containers have popularised the concept of read-only/immutable infrastructure and lead to changes in system and application architecture across the IT industry. However nearly every application generates some data that will need to persist long after the life-span of the container that generated it. This talk will look at the best practices around persistent storage with containers, from providing design advice around the construction of your application/container to the functionality provided from storage vendors through the Docker Volume driver plugins.
Cloud Foundry and OpenStack – Marriage Made in Heaven !Animesh Singh
Cloud Foundry Summit 2014 Presentation: Bring the world's best IaaS to the world's best PaaS, In this talk IBM and Rackspace are going to share their experiences of running Cloud Foundry on OpenStack. The talk will focus on how CloudFoundry and OpenStack complement each other, how they technically integrate using Cloud provider interface (CPI), how could we automate OpenStack setup for Cloud Foundry deployments, and what are some of the best practices for configuring a scalable environment.
Beyond Ingresses - Better Traffic Management in KubernetesMark McBride
Kubernetes makes deploying code easy, but conflating deploys and releases is risky. Using smarter proxies you can dramatically reduce the risk of a release, which in turn helps you ship code to customers faster.
Docker storage designing a platform for persistent dataDocker, Inc.
Docker containers have popularised the concept of read-only/immutable infrastructure and lead to changes in system and application architecture across the IT industry. However nearly every application generates some data that will need to persist long after the life-span of the container that generated it. This talk will look at the best practices around persistent storage with containers, from providing design advice around the construction of your application/container to the functionality provided from storage vendors through the Docker Volume driver plugins.
Cloud Foundry and OpenStack – Marriage Made in Heaven !Animesh Singh
Cloud Foundry Summit 2014 Presentation: Bring the world's best IaaS to the world's best PaaS, In this talk IBM and Rackspace are going to share their experiences of running Cloud Foundry on OpenStack. The talk will focus on how CloudFoundry and OpenStack complement each other, how they technically integrate using Cloud provider interface (CPI), how could we automate OpenStack setup for Cloud Foundry deployments, and what are some of the best practices for configuring a scalable environment.
Delivering Container-based Apps to IoT Edge devicesAjeet Singh Raina
I presented it during Dockercon. This talk was all about AI + Docker + IoT. Showcased how Docker app talk to Sensors, GPUs and Camera module and demo'ed how sensors data can be visualized over Grafana dashboard - all running on a IoT Edge device.
OSDC 2018 | Lifecycle of a resource. Codifying infrastructure with Terraform ...NETWAYS
Immutable infrastructure is a way to success, but what about the lifecycle of individual resources. This talk is about evolution of resources, code structure, Terraform coding tricks, composition and refactoring.
DCEU 18: Use Cases and Practical Solutions for Docker Container Storage on Sw...Docker, Inc.
Mark Church - Product Manager, Docker
Don Stewart - Solutions Architect, Docker
Persistent storage has quickly advanced from something considered incompatible with containers to a mature set of solutions and patterns that have been thoroughly adopted by the industry. We’ll define the persistent characteristics of different use-cases and map these to some of the many solutions that exist for container storage. From this talk you’ll learn about the storage options available to users on Swarm, Kubernetes, on-premises, cloud, and how they work and compare to each other. You’ll also learn how to characterize different persistent application requirements and the solutions best for suited for them.
Axel Koehler from Nvidia presented this deck at the 2016 HPC Advisory Council Switzerland Conference.
“Accelerated computing is transforming the data center that delivers unprecedented through- put, enabling new discoveries and services for end users. This talk will give an overview about the NVIDIA Tesla accelerated computing platform including the latest developments in hardware and software. In addition it will be shown how deep learning on GPUs is changing how we use computers to understand data.”
In related news, the GPU Technology Conference takes place April 4-7 in Silicon Valley.
Watch the video presentation: http://insidehpc.com/2016/03/tesla-accelerated-computing/
See more talks in the Swiss Conference Video Gallery:
http://insidehpc.com/2016-swiss-hpc-conference/
Sign up for our insideHPC Newsletter:
http://insidehpc.com/newsletter
Building analytical microservices powered by jupyter kernelsLuciano Resende
The Jupyter Kernels, which abstracts the computing engine used in Jupyter Notebooks, are a very powerful component that can be reutilized in different scenarios to bring analytical capabilities to applications. In this session, we will discuss how you can build a simple python based micro service that leverages Jupyter Kernels to incorporate sentiment analysis to the service it provides.
Feeling overwhelmed while getting started with containers? Have you been tasked to figure out how to train everyone back at your organization? There's just so much to learn and teach! In this talk, we'll start with a tiny bit of history to motivate the "why" and quickly move into the "what" by explaining what container and images actually are (they're not just magical black boxes!). We'll talk about how volumes help with data persistence and include an overview of Docker Compose and even orchestration. There will be plenty of live demos and fun!
An overview of Mesos and Kubernetes ecosystem including overview, architecture, customers and partners. For a beginner it will give a good covering of all the basics!
Scaling notebooks for Deep Learning workloadsLuciano Resende
Deep learning workloads are computing intensive, and training these type of models is better done with specialized hardware like GPUs. Luciano Resende outlines a pattern for building deep learning models using the Jupyter Notebook’s interactive development in commodity hardware and leveraging platforms and services such as Fabric for Deep Learning (FfDL) for cost-effective full dataset training of deep learning models.
DCEU 18: Automating Docker Enterprise: Hands-off Install and UpgradeDocker, Inc.
Brett Inman - Manager, Infrastructure Engineering, Docker
Loke Norlin Johannessen - Senior System Specialist, Alm. Brand
Sune Keller - IT Architect, Alm. Brand
At both Alm. Brand and Docker, we’ve been running Docker Enterprise in production since the first beta. Both teams have gathered several learnings that were ripe for being codified, both to save time and provide certainty in daily operations. This is the story of how each company moved to a more automated and declarative approach. In this talk, we will share how Alm. Brand and Docker use tools like Packer, Terraform, Ansible, Cloudformation, Salt, and GitLab to build and upgrade Docker Enterprise clusters, comparing how both companies have solved for automating the Docker Enterprise infrastructure lifecycle. This approach focuses on immutable or disposable infrastructure, software components, and most importantly the processes to help you build your own automation solution for Docker Enterprise to fit your infrastructure needs.
Kubernetes Concepts And Architecture Powerpoint Presentation SlidesSlideTeam
Get these visually appealing Kubernetes Concepts And Architecture PowerPoint Presentation Slides to discuss the process of operating containerized applications. You can display the need for containers by the company with the help of an open-source architecture PPT slideshow. The architecture of containers can be demonstrated with the help of a visually appealing PPT slideshow. The reasons for opting for Kubernetes by an organization can be explained to your teammates with the help of containers PowerPoint infographics. Highlight the roadmap for installing Kubernetes in the organization by using content-ready PPT slides. Take the assistance of visually appealing PPT templates to depict the major advantages of Kubernetes such as improving productivity, the stability of application run, and many more. After that, display 30 60 90 days plan to implement Kubernetes in the organization. Display the key components of Kubernetes with the help of a diagram using this professionally designed cluster architecture PPT layouts. Describe the functionality of each components of Kubernetes. Hence, download Kubernetes architecture PPT slides to easily and efficiently manage the clusters. https://bit.ly/34DWa7x
Wouldn't it be great for a new developer on your team to have their dev environment totally set up on their first day? What about having your CI tests running in the background while you work on new features? What about having the confidence that your dev environment mirrors testing and prod? Containers enable this to become reality, along with other great benefits like keeping dependencies nice and tidy and making packaged code easier to share. Come learn about the ways containers can help you build and ship software easily.
The slideshow gives an introduction and overview of APIs and its growth, importance in the cloud and mobile era of computing. It also briefs about various business models and the API management services / platforms available.
Delivering Container-based Apps to IoT Edge devicesAjeet Singh Raina
I presented it during Dockercon. This talk was all about AI + Docker + IoT. Showcased how Docker app talk to Sensors, GPUs and Camera module and demo'ed how sensors data can be visualized over Grafana dashboard - all running on a IoT Edge device.
OSDC 2018 | Lifecycle of a resource. Codifying infrastructure with Terraform ...NETWAYS
Immutable infrastructure is a way to success, but what about the lifecycle of individual resources. This talk is about evolution of resources, code structure, Terraform coding tricks, composition and refactoring.
DCEU 18: Use Cases and Practical Solutions for Docker Container Storage on Sw...Docker, Inc.
Mark Church - Product Manager, Docker
Don Stewart - Solutions Architect, Docker
Persistent storage has quickly advanced from something considered incompatible with containers to a mature set of solutions and patterns that have been thoroughly adopted by the industry. We’ll define the persistent characteristics of different use-cases and map these to some of the many solutions that exist for container storage. From this talk you’ll learn about the storage options available to users on Swarm, Kubernetes, on-premises, cloud, and how they work and compare to each other. You’ll also learn how to characterize different persistent application requirements and the solutions best for suited for them.
Axel Koehler from Nvidia presented this deck at the 2016 HPC Advisory Council Switzerland Conference.
“Accelerated computing is transforming the data center that delivers unprecedented through- put, enabling new discoveries and services for end users. This talk will give an overview about the NVIDIA Tesla accelerated computing platform including the latest developments in hardware and software. In addition it will be shown how deep learning on GPUs is changing how we use computers to understand data.”
In related news, the GPU Technology Conference takes place April 4-7 in Silicon Valley.
Watch the video presentation: http://insidehpc.com/2016/03/tesla-accelerated-computing/
See more talks in the Swiss Conference Video Gallery:
http://insidehpc.com/2016-swiss-hpc-conference/
Sign up for our insideHPC Newsletter:
http://insidehpc.com/newsletter
Building analytical microservices powered by jupyter kernelsLuciano Resende
The Jupyter Kernels, which abstracts the computing engine used in Jupyter Notebooks, are a very powerful component that can be reutilized in different scenarios to bring analytical capabilities to applications. In this session, we will discuss how you can build a simple python based micro service that leverages Jupyter Kernels to incorporate sentiment analysis to the service it provides.
Feeling overwhelmed while getting started with containers? Have you been tasked to figure out how to train everyone back at your organization? There's just so much to learn and teach! In this talk, we'll start with a tiny bit of history to motivate the "why" and quickly move into the "what" by explaining what container and images actually are (they're not just magical black boxes!). We'll talk about how volumes help with data persistence and include an overview of Docker Compose and even orchestration. There will be plenty of live demos and fun!
An overview of Mesos and Kubernetes ecosystem including overview, architecture, customers and partners. For a beginner it will give a good covering of all the basics!
Scaling notebooks for Deep Learning workloadsLuciano Resende
Deep learning workloads are computing intensive, and training these type of models is better done with specialized hardware like GPUs. Luciano Resende outlines a pattern for building deep learning models using the Jupyter Notebook’s interactive development in commodity hardware and leveraging platforms and services such as Fabric for Deep Learning (FfDL) for cost-effective full dataset training of deep learning models.
DCEU 18: Automating Docker Enterprise: Hands-off Install and UpgradeDocker, Inc.
Brett Inman - Manager, Infrastructure Engineering, Docker
Loke Norlin Johannessen - Senior System Specialist, Alm. Brand
Sune Keller - IT Architect, Alm. Brand
At both Alm. Brand and Docker, we’ve been running Docker Enterprise in production since the first beta. Both teams have gathered several learnings that were ripe for being codified, both to save time and provide certainty in daily operations. This is the story of how each company moved to a more automated and declarative approach. In this talk, we will share how Alm. Brand and Docker use tools like Packer, Terraform, Ansible, Cloudformation, Salt, and GitLab to build and upgrade Docker Enterprise clusters, comparing how both companies have solved for automating the Docker Enterprise infrastructure lifecycle. This approach focuses on immutable or disposable infrastructure, software components, and most importantly the processes to help you build your own automation solution for Docker Enterprise to fit your infrastructure needs.
Kubernetes Concepts And Architecture Powerpoint Presentation SlidesSlideTeam
Get these visually appealing Kubernetes Concepts And Architecture PowerPoint Presentation Slides to discuss the process of operating containerized applications. You can display the need for containers by the company with the help of an open-source architecture PPT slideshow. The architecture of containers can be demonstrated with the help of a visually appealing PPT slideshow. The reasons for opting for Kubernetes by an organization can be explained to your teammates with the help of containers PowerPoint infographics. Highlight the roadmap for installing Kubernetes in the organization by using content-ready PPT slides. Take the assistance of visually appealing PPT templates to depict the major advantages of Kubernetes such as improving productivity, the stability of application run, and many more. After that, display 30 60 90 days plan to implement Kubernetes in the organization. Display the key components of Kubernetes with the help of a diagram using this professionally designed cluster architecture PPT layouts. Describe the functionality of each components of Kubernetes. Hence, download Kubernetes architecture PPT slides to easily and efficiently manage the clusters. https://bit.ly/34DWa7x
Wouldn't it be great for a new developer on your team to have their dev environment totally set up on their first day? What about having your CI tests running in the background while you work on new features? What about having the confidence that your dev environment mirrors testing and prod? Containers enable this to become reality, along with other great benefits like keeping dependencies nice and tidy and making packaged code easier to share. Come learn about the ways containers can help you build and ship software easily.
The slideshow gives an introduction and overview of APIs and its growth, importance in the cloud and mobile era of computing. It also briefs about various business models and the API management services / platforms available.
If you're like most of the world, you're on an aggressive race to implement machine learning applications and on a path to get to deep learning. If you can give better service at a lower cost, you will be the winners in 2030. But infrastructure is a key challenge to getting there. What does the technology infrastructure look like over the next decade as you move from Petabytes to Exabytes? How are you budgeting for more colossal data growth over the next decade? How do your data scientists share data today and will it scale for 5-10 years? Do you have the appropriate security, governance, back-up and archiving processes in place? This session will address these issues and discuss strategies for customers as they ramp up their AI journey with a long term view.
The latest distributed system utilizing the cloud is a very complicated configuration in which the components span a plurality of components. Applications for customers are part of products, and service quality targets directly linked to business indicators are needed. Legacy monitoring system based on traditional system management is not linked not only to business indicators but also to measure service quality. Google advocates the idea of site reliability engineering (SRE) and introduces efforts to measure quality of service. Based on the concept of SRE, the service quality monitoring system collects and analyzes logs from various components not only application codes but also whole infrastructure components. Since very large amounts of data must be processed in real time, it is necessary to design carefully with reference to the big data architecture. To utilize this system, you can measure the quality of service, and make it possible to continuously improve the service quality.
Scaling out Driverless AI with IBM Spectrum Conductor - Kevin Doyle - H2O AI ...Sri Ambati
This talk was recorded in London on Oct 30, 2018 and can be viewed here: https://youtu.be/lk2NXurrwAA
This talk highlights the integration of Driverless AI with IBM Spectrum Conductor. The integration demonstrates how you can deploy, manage, and scale out to have multiple Driverless AI instances running within your cluster per user to help maximize the efficiency and security of the cluster. The integration includes failover for Driverless AI instances, so that users can continue to work without needing to find another host to start Driverless AI on. In addition, the integration of H2O Sparkling Water with IBM Spectrum Conductor as a notebook is highlighted; as well as the benefits of running H20 Sparkling water within the cluster to maximize your cluster utilization across different workloads.For both Driverless AI and H2O Sparkling Water, a demo will be provided and a future plan for the integrations is highlighted.
Bio: Kevin Doyle is the lead architect of IBM Spectrum Conductor at IBM, where he works with customers to deploy and manage all workloads; especially Spark and deep learning workloads to on-premise clusters. Kevin has been working on distributed computing, grid, cloud, and big data for the past five years with a focus on the management and lifecycle of workloads.
How to optimize Hortonworks Apache Spark ML workloads on Power - POWER 8/9 architecture is the latest offering from IBM and OpenPower foundation. It is the perfect platform for optimizing Hortonworks Spark's performance. During this presentation we will walk the audience through steps required to optimize YARN, HDFS, and Spark on a Power cluster.
Step required:
1) Classify workload into CPU, Memory, IO or mixed (CPU, memory, IO) intensive
2) Characterize "out-of-box" Hortonworks spark workload to understand CPU, Memory, IO and Network performance characteristics
3) Floor Plan cluster resources
4) Tune "out-of-box" workload to navigate "Roofline" Performance space in the above named dimensions
5) If workload is Memory / IO/ Network intensive bound then tune SPARK to increase operational intensity operations/byte as much as possible to make it CPU bound
6) Divide search space into regions and perform exhaustive search.
7) Identify Performance bottlenecks by resource monitoring and tune the System, JVM or application layer by profiling application and hardware counters if required.
OS for AI: Elastic Microservices & the Next Gen of MLNordic APIs
AI has been a hot topic lately, with advances being made constantly in what is possible, there has not been as much discussion of the infrastructure and scaling challenges that come with it. How do you support dozens of different languages and frameworks, and make them interoperate invisibly? How do you scale to run abstract code from thousands of different developers, simultaneously and elastically, while maintaining less than 15ms of overhead?
At Algorithmia, we’ve built, deployed, and scaled thousands of algorithms and machine learning models, using every kind of framework (from scikit-learn to tensorflow). We’ve seen many of the challenges faced in this area, and in this talk I’ll share some insights into the problems you’re likely to face, and how to approach solving them.
In brief, we’ll examine the need for, and implementations of, a complete “Operating System for AI” – a common interface for different algorithms to be used and combined, and a general architecture for serverless machine learning which is discoverable, versioned, scalable and sharable.
Thomas Weise, Apache Apex PMC Member and Architect/Co-Founder, DataTorrent - ...Dataconomy Media
Thomas Weise, Apache Apex PMC Member and Architect/Co-Founder of DataTorrent presented "Streaming Analytics with Apache Apex" as part of the Big Data, Berlin v 8.0 meetup organised on the 14th of July 2016 at the WeWork headquarters.
Webinar: High Performance MongoDB Applications with IBM POWER8MongoDB
Innovative companies are building Internet of Things, mobile, content management, single view, and big data apps on top of MongoDB. In this session, we'll explore how the IBM POWER8 platform brings new levels of performance and ease of configuration to these solutions which already benefit from easier and faster design and development using MongoDB.
2689 - Exploring IBM PureApplication System and IBM Workload Deployer Best Pr...Hendrik van Run
IBM IMPACT 2013 presentation
This lecture will provide an overview of a combination of design, development, configuration and deployment best practices for IBM PureApplication System and IBM Workload Deployer captured from customer engagement experiences.
Presenting the newest version of Cloudify - 4.6 including a orchestrated SD-WAN demo from MEF18 where Cloudify is used as the orchestration platform for uCPE based on containers.
Modernizing Testing as Apps Re-ArchitectDevOps.com
Applications are moving to cloud and containers to boost reliability and speed delivery to production. However, if we use the same old approaches to testing, we'll fail to achieve the benefits of cloud. But what do we really need to change? We know we need to automate tests, but how do we keep our automation assets from becoming obsolete? Automatically provisioning test environments seems close, but some parts of our applications are hard to move to cloud.
Migrating Mission-Critical Workloads to Intel ArchitectureIntel IT Center
Based on Intel's RISC/UNIX Migration Planning Guide, this powerpoint can be used to simplify your RISC/UNIX* migration to Intel® Xeon® processor-based solutions running Linux* or Windows* operating systems. You’ll gain practical guidance, including the steps needed to create a solid project plan.
Similar to Lessons Learned from Deploying Apache Spark as a Service on IBM Power Systems in the Cloud (20)
DevOps and Testing slides at DASA ConnectKari Kakkonen
My and Rik Marselis slides at 30.5.2024 DASA Connect conference. We discuss about what is testing, then what is agile testing and finally what is Testing in DevOps. Finally we had lovely workshop with the participants trying to find out different ways to think about quality and testing in different parts of the DevOps infinity loop.
Kubernetes & AI - Beauty and the Beast !?! @KCD Istanbul 2024Tobias Schneck
As AI technology is pushing into IT I was wondering myself, as an “infrastructure container kubernetes guy”, how get this fancy AI technology get managed from an infrastructure operational view? Is it possible to apply our lovely cloud native principals as well? What benefit’s both technologies could bring to each other?
Let me take this questions and provide you a short journey through existing deployment models and use cases for AI software. On practical examples, we discuss what cloud/on-premise strategy we may need for applying it to our own infrastructure to get it to work from an enterprise perspective. I want to give an overview about infrastructure requirements and technologies, what could be beneficial or limiting your AI use cases in an enterprise environment. An interactive Demo will give you some insides, what approaches I got already working for real.
UiPath Test Automation using UiPath Test Suite series, part 4DianaGray10
Welcome to UiPath Test Automation using UiPath Test Suite series part 4. In this session, we will cover Test Manager overview along with SAP heatmap.
The UiPath Test Manager overview with SAP heatmap webinar offers a concise yet comprehensive exploration of the role of a Test Manager within SAP environments, coupled with the utilization of heatmaps for effective testing strategies.
Participants will gain insights into the responsibilities, challenges, and best practices associated with test management in SAP projects. Additionally, the webinar delves into the significance of heatmaps as a visual aid for identifying testing priorities, areas of risk, and resource allocation within SAP landscapes. Through this session, attendees can expect to enhance their understanding of test management principles while learning practical approaches to optimize testing processes in SAP environments using heatmap visualization techniques
What will you get from this session?
1. Insights into SAP testing best practices
2. Heatmap utilization for testing
3. Optimization of testing processes
4. Demo
Topics covered:
Execution from the test manager
Orchestrator execution result
Defect reporting
SAP heatmap example with demo
Speaker:
Deepak Rai, Automation Practice Lead, Boundaryless Group and UiPath MVP
Epistemic Interaction - tuning interfaces to provide information for AI supportAlan Dix
Paper presented at SYNERGY workshop at AVI 2024, Genoa, Italy. 3rd June 2024
https://alandix.com/academic/papers/synergy2024-epistemic/
As machine learning integrates deeper into human-computer interactions, the concept of epistemic interaction emerges, aiming to refine these interactions to enhance system adaptability. This approach encourages minor, intentional adjustments in user behaviour to enrich the data available for system learning. This paper introduces epistemic interaction within the context of human-system communication, illustrating how deliberate interaction design can improve system understanding and adaptation. Through concrete examples, we demonstrate the potential of epistemic interaction to significantly advance human-computer interaction by leveraging intuitive human communication strategies to inform system design and functionality, offering a novel pathway for enriching user-system engagements.
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Jeffrey Haguewood
Sidekick Solutions uses Bonterra Impact Management (fka Social Solutions Apricot) and automation solutions to integrate data for business workflows.
We believe integration and automation are essential to user experience and the promise of efficient work through technology. Automation is the critical ingredient to realizing that full vision. We develop integration products and services for Bonterra Case Management software to support the deployment of automations for a variety of use cases.
This video focuses on the notifications, alerts, and approval requests using Slack for Bonterra Impact Management. The solutions covered in this webinar can also be deployed for Microsoft Teams.
Interested in deploying notification automations for Bonterra Impact Management? Contact us at sales@sidekicksolutionsllc.com to discuss next steps.
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...UiPathCommunity
💥 Speed, accuracy, and scaling – discover the superpowers of GenAI in action with UiPath Document Understanding and Communications Mining™:
See how to accelerate model training and optimize model performance with active learning
Learn about the latest enhancements to out-of-the-box document processing – with little to no training required
Get an exclusive demo of the new family of UiPath LLMs – GenAI models specialized for processing different types of documents and messages
This is a hands-on session specifically designed for automation developers and AI enthusiasts seeking to enhance their knowledge in leveraging the latest intelligent document processing capabilities offered by UiPath.
Speakers:
👨🏫 Andras Palfi, Senior Product Manager, UiPath
👩🏫 Lenka Dulovicova, Product Program Manager, UiPath
Securing your Kubernetes cluster_ a step-by-step guide to success !KatiaHIMEUR1
Today, after several years of existence, an extremely active community and an ultra-dynamic ecosystem, Kubernetes has established itself as the de facto standard in container orchestration. Thanks to a wide range of managed services, it has never been so easy to set up a ready-to-use Kubernetes cluster.
However, this ease of use means that the subject of security in Kubernetes is often left for later, or even neglected. This exposes companies to significant risks.
In this talk, I'll show you step-by-step how to secure your Kubernetes cluster for greater peace of mind and reliability.
Generating a custom Ruby SDK for your web service or Rails API using Smithyg2nightmarescribd
Have you ever wanted a Ruby client API to communicate with your web service? Smithy is a protocol-agnostic language for defining services and SDKs. Smithy Ruby is an implementation of Smithy that generates a Ruby SDK using a Smithy model. In this talk, we will explore Smithy and Smithy Ruby to learn how to generate custom feature-rich SDKs that can communicate with any web service, such as a Rails JSON API.
Accelerate your Kubernetes clusters with Varnish CachingThijs Feryn
A presentation about the usage and availability of Varnish on Kubernetes. This talk explores the capabilities of Varnish caching and shows how to use the Varnish Helm chart to deploy it to Kubernetes.
This presentation was delivered at K8SUG Singapore. See https://feryn.eu/presentations/accelerate-your-kubernetes-clusters-with-varnish-caching-k8sug-singapore-28-2024 for more details.
Monitoring Java Application Security with JDK Tools and JFR Events
Lessons Learned from Deploying Apache Spark as a Service on IBM Power Systems in the Cloud
1. Lessons Learned from Deploying Apache Spark
as a Service on IBM Power Systems in the Cloud
Indrajit (I.P) Poddar, STSM, IBM Systems Technical Strategy
Randy Swanberg, DE, IBM Power Systems Software and Solutions
2. Please Note:
1
• IBM’s statements regarding its plans, directions, and intent are subject to change or withdrawal without notice at IBM’s sole
discretion.
• Information regarding potential future products is intended to outline our general product direction and it should not be relied on in
making a purchasing decision.
• The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any
material, code or functionality. Information about potential future products may not be incorporated into any contract.
• The development, release, and timing of any future features or functionality described for our products remains at our sole discretion.
• Performance is based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual
throughput or performance that any user will experience will vary depending upon many factors, including considerations such as the
amount of multiprogramming in the user’s job stream, the I/O configuration, the storage configuration, and the workload processed.
Therefore, no assurance can be given that an individual user will achieve results similar to those stated here.
3. Agenda
• Infrastructure considerations for a differentiated cloud data service
• Apache Spark – a popular big data framework
• Lessons learned on an alternative infrastructure with OpenPOWER systems
1. Open source stack for cloud native agile development
2. Management stack for automation and continuous integration
3. Efficient resource allocation and scheduling for multi-tenancy
4. Cloud infrastructure economics
5. Potential of acceleration under the cloud service
• Putting it all together
• Summary / Questions
2
4. Infrastructure for a differentiated Cloud Data Service
1. Agile Open Source development
experience
• Dynamic and flexible provisioning and
management
• Automated deployment and continuous
integration
2. Cost effective high performance
server infrastructure
3. Economical cloud storage service
with encryption
3
5. Big Data in the Cloud
Example use cases, architectures and components..
6. Big Data Journey
5
Operations Data
Warehouse
Insight Inspired
Decision Making
Insight Driven
Business
Transformation
Value
Big-Data Maturity
• Cheaper Storage
• Data Lake
• ETL Offload
• Cold Data Offload
• Queryable Archive
• Full Data Analysis (not
just samples)
• Extract Value from
non-relational data
• View of all enterprise
data
• Exploratory Analysis
and Discovery
• New Business Models
• Real time risk aware
decision making
• Real time fraud and
threat detection
• Optimize operations
• Attract and Retain
Customers
Most are somewhere here
7. Demand for Business Value from ALL Data Sources
Transaction and
application data
Machine,
sensor data
Enterprise
content
Image, geospatial,
video
Social data
Third-party data
Deep
Analytics
data zone
EDW and
data mart
zone
New Customer
Insights
Discover
Relationship
Risk Aware
Decisions
Early Warnings
New Business
Opportunities
Fraud Detection
8. What is Apache Spark?
• Unified Analytics Platform
– Combine streaming, graph, machine learning
and sql analytics on a single platform
– Simplified, multi-language programming model
– Interactive and Batch
• In-Memory Design
– Pipelines multiple iterations on single copy of
data in memory
– Superior Performance
– Natural Successor to MapReduce
7
Fast and general engine for
large-scale data processing
Spark Core API
R Scala SQL Python Java
Spark SQL Streaming MLlib GraphX
9. Anatomy of Apache Spark as a
Cloud Service
Stack components, web services, continuous
integration..
10. Interactive iPython notebook with Matplotlib GUI
9
Host interactive analytic apps on ipython server to ease code sharing and reuse
11. Architectural components of a Apache Spark Cloud Service
10
Object Store for dataPlatform as a Service
Multi-tenant Spark drivers
and executors
Multi-tenant interactive Jupyter
ipython notebook servers with
matplotlib GUI
Shared Compute Cluster
Infrastructure as a Service
Bare metals Virtual Machines
Continuous Integration
Build Deploy
Shared storage IBM Spectrum Scale (GPFS)
Test
12. A prototype deployment on POWER systems
11
Goal: Production deployment
IBM Power System S812LC and S822LC, Tyan OpenPOWER
Development environment
Continuous Integration
Deployment Automation
Bare metals
Virtual Machines
Docker Containers
Docker containers
13. Lessons Learned
Open source stack, efficient resource allocation and continuous
integration, better economics and potential for acceleration on
OpenPOWER systems ..
15. OpenStack on POWER for Dev-Test environments
14
We used IBM Cloud Manager version 4.3 with OpenStack Kilo on PowerKVM 2.1.1
IBM Cloud Orchestrator is another option
16. Continuous Integration with IBM Urban Code, OpenStack and Docker
targeting POWER systems
15
Create multiple deployment and development environments and visual deployment processes in IBM Urban Code Deployment
Run only UCD agents and relay on POWER VMs or bare metals
17. Continuous Integration Flows
16
Urban Code
Deploy
Server (x86)
Git Server
(x86)
Asset
Repository
(x86)
Dev-Test Env
(OpenPOWER OpenStack VMs)
Build Env
(POWER Docker Containers)
Future Production Env
(OpenPOWER Bare metals)
1. Check in automation code
2. Build artifacts in Docker
3. Store built artifacts in a repository
4. Pull in artifacts into deployment automation
5. Deploy artifacts into dev-test env
6. Deploy artifacts into prod env
18. Efficient resource allocation using Platform Symphony
17
• Share system resources (CPU, memory)
with a distributed scheduler
• Platform Symphony with Application
Service Controller (ASC) V7.1.1 and the
EGO scheduler for Ubuntu 14.04.2 on
POWER
• Platform Symphony + ASC features
• Fine-grained scheduling
• Resource reclaim
• Standby service
• Data locality
• Shared storage through IBM Spectrum
Scale™
19. Economics of the Server Infrastructure
• Attributes directly influencing the economics
of hosting a Cloud Service:
• Number of servers needed to deliver a
competitive quality of service and response
time
• Cost of the individual servers or rental
• Number of users and jobs that can be
hosted concurrently (multi-tenancy)
• End user price charged for the service
• All of which are directly impacted by the
performance of the Server
18
20. Measuring Performance of Spark on POWER
19
The following charts show Performance
results of comparing multiple Spark Workloads
from SparkBench using data sizes from
100GB to 10TB
(https://github.com/SparkTC/spark-bench)
7-node cluster of Intel Haswell servers
• E5-2620 V3
• 12-core
• 256GB
vs
7-node cluster of POWER servers
• POWER8 S812LC
• 10-core
• 256GB
• Machine Learning (Spark MLlib)
• Matrix Factorization
• Logistic Regression
• Support Vector Machine
• SQL (Spark SQL)
sqlContext.sql("SELECT COUNT(*) FROM orderTab").count()
sqlContext.sql("SELECT COUNT(*) FROM orderTab where bid>5000").count()
sqlContext.sql("SELECT * FROM oitemTab WHERE price>250").count()
sqlContext.sql("SELECT * FROM oitemTab WHERE price>500").count()
sqlContext.sql("SELECT * FROM orderTab r JOIN oitemTab s ON r.oid =
s.oid").count()
• Graph (Spark GraphX)
• Page Rank
• Triangle Count
• Singular Value Decomp++
21. System Performance of Spark on POWER
20
Machine Learning SQL Graph
1.7X
Raw System Performance
Options for the Service Provider:
• Deliver higher qualities of service
• 70% faster job completion
times on average
• Faster time-to-insight
• Charge higher premium for the service
• Competitive advantage for the service
22. 21
Per Core Performance of Spark on POWER
Per-core Performance View
• 70 POWER8 Cores vs. 84 Intel Cores
• Enables headroom for better system
resource utilization
Machine Learning SQL Graph
2X
23. 22
Price Performance of Spark on POWER
Machine Learning SQL Graph
1.5X
Price Performance View*
Options for the Service Provider:
• Spend 33% less on infrastructure
supporting the same amount of workload
• Spend the same on infrastructure but
host 50% more workload
• Lower the price for the service for
competitive advantage
* - based on preliminary SoftLayer pricing targets – subject to change
24. But wait….there’s more to the story
23
0
0.5
1
1.5
2
2.5
3
E5-2620
v3100GB
M
at.Fact.
100GB
(in
m
em
)LR
1TB
(in
m
em
)LR
1TB
(50/50)LR1TB
SVM
10TB
LR
1TB
5
query
2TB
5
query130GB
Page
Rank
1TB
Triangle
Cnt
1TB
SVD++
AVERAGE
RelativeSystemPerformance
Spark Workloads
• A Deeper Look at the System
Performance profile for one of the
workloads close to our overall
average relative performance
• Machine Learning Logistic
Regression on a 1TB data set that
had a relative performance of 1.74X
Machine Learning SQL Graph
1.7X
25. More efficient use of resources (Spark 1TB Logistic Regression Example)
24
0
20
40
60
80
100
22:44
22:44
22:45
22:45
22:46
22:46
22:47
22:47
22:48
22:48
22:49
22:49
22:50
22:50
22:51
22:51
22:52
22:52
22:53
22:53
22:54
22:54
22:55
CPU POWER
User% Sys% Wait%
0
20
40
60
80
100
10:14
10:14
10:15
10:16
10:16
10:17
10:18
10:18
10:19
10:20
10:20
10:21
10:22
10:22
10:23
10:24
10:24
10:25
10:26
10:26
10:27
10:28
10:28
CPU Haswell
User% Sys% Wait%
-1500
-1000
-500
0
500
1000
22:44
22:44
22:45
22:45
22:46
22:46
22:47
22:47
22:48
22:48
22:49
22:49
22:50
22:50
22:51
22:51
22:52
22:52
22:53
22:53
22:54
22:54
22:55
MB/sec
Network I/O POWER
Total-Read Total-Write (-ve)
-600
-500
-400
-300
-200
-100
0
100
200
300
400
10:14
10:14
10:15
10:15
10:16
10:17
10:17
10:18
10:18
10:19
10:19
10:20
10:21
10:21
10:22
10:22
10:23
10:24
10:24
10:25
10:25
10:26
10:26
10:27
10:28
10:28
MB/sec
Network I/O Haswell
Total-Read Total-Write (-ve)
POWER
• CPU headroom to
host higher density
• More data pushed
over network due to
higher thread density
Haswell
• CPU fully pegged on
just this workload
• Underutilizing the
Network Resource
26. 0
50000
100000
150000
200000
250000
300000
350000
400000
Runtime(ms)
Total Heap Memory
x Degrees of Separation on
Spark
Disk
CAPI/Flash
25
CAPI Flash for RDD Cache = 4X memory
reduction at equal performance
Next Steps - Acceleration in the Cloud
RDMA for Spark Shuffle = 30% Better
Response Time, Lower CPU Utilization
• CAPI Flash and RDMA can be Leveraged Transparently to Spark Applications under the Cloud Service
• Coming…. HDFS CAPI FPGA Erasure Code Acceleration, CAPI FPGA Compression Acceleration, ….
27. 26
Acceleration of Spark with GPUs:
• Adverse Drug Reaction Prediction built on Spark
• 25X Speed up for Building Model stage (using Spark Mllib Logistic Regression)
• Again, Transparent to the Spark Application
• Game changer for Personalized Medicine
28. More efficient, cost effective, balanced cloud resources
• Better quality of service through workload acceleration and real time insights
• Efficient scale out architecture avoiding imbalanced resources
• New controls to balance resource utilization
27
GPUs and FPGAs for Compute
offload, consolidation, specialized
acceleration
CAPI Flash for Memory
consolidation/expansion, and Storage
acceleration
RDMA for better latency, better
network utilization and lower CPU
utilization
30. Putting it All Together….
29
Object Store for dataPlatform as a Service
Multi-tenant Spark drivers
and executors
Multi-tenant interactive Jupyter
ipython notebook servers with
matplotlib GUI
Shared Compute Cluster
Infrastructure as a Service
Bare metals Virtual Machines
Continuous Integration
Build Deploy
Shared storage IBM Spectrum Scale (GPFS)
Test
Docker
31. Summary
30
• Big Data solutions in the Cloud demand elasticity and scale
• Real time insights from all sources of data will become the norm
• Try open source Apache Spark with IBM Platform Symphony
• OpenPOWER systems can differentiate your cloud data service through:
• Improved cloud infrastructure economics and cost performance advantage
• An agile open source development experience
• Advanced forms of acceleration in cloud infrastructures will further differentiate
services
33. Notices and Disclaimers Con’t.
32
Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not
tested those products in connection with this publication and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products.
Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM does not warrant the quality of any third-party products, or the
ability of any such third-party products to interoperate with IBM’s products. IBM EXPRESSLY DISCLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING BUT
NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
The provision of the information contained h erein is not intended to, and does not, grant any right or license under any IBM patents, copyrights, trademarks or other intellectual
property right.
IBM, the IBM logo, ibm.com, Aspera®, Bluemix, Blueworks Live, CICS, Clearcase, Cognos®, DOORS®, Emptoris®, Enterprise Document Management System™, FASP®,
FileNet®, Global Business Services ®, Global Technology Services ®, IBM ExperienceOne™, IBM SmartCloud®, IBM Social Business®, Information on Demand, ILOG,
Maximo®, MQIntegrator®, MQSeries®, Netcool®, OMEGAMON, OpenPower, PureAnalytics™, PureApplication®, pureCluster™, PureCoverage®, PureData®,
PureExperience®, PureFlex®, pureQuery®, pureScale®, PureSystems®, QRadar®, Rational®, Rhapsody®, Smarter Commerce®, SoDA, SPSS, Sterling Commerce®,
StoredIQ, Tealeaf®, Tivoli®, Trusteer®, Unica®, urban{code}®, Watson, WebSphere®, Worklight®, X-Force® and System z® Z/OS, are trademarks of International Business
Machines Corporation, registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM
trademarks is available on the Web at "Copyright and trademark information" at: www.ibm.com/legal/copytrade.shtml.
34. Thank You
Your Feedback is Important!
Access the InterConnect 2016 Conference Attendee
Portal to complete your session surveys from your
smartphone,
laptop or conference kiosk.