It’s been a while since my last post. A lot has happened, and I have been seriously focussing on VMworld. As you may know, VMworld 2020 is fully digital and online, and best of all: it’s completely FREE! I submitted just a single session on integrating Bitfusion and Kubernetes, but suddenly some opportunities arose and now I have six sessions in the catalog. In this post, I will focus on both these sessions as well as some other sessions I look forward to.
My own sessions
Like mentioned above, I got six session in the VMworld agenda. Four of them are semi-live sessions, and two of them are on-demand. The four live sessions can be scheduled at this moment, and the two on-demand ones will be available during VMworld to watch whenever it suits you. You can use the following link to view them all:
Using Kubernetes and vSphere Bitfusion for ML Workloads [HCP1266]
VMware vSphere 7 is the perfect platform to run machine learning (ML) and deep learning (DL) workloads. But why? Data science departments push the limits of their compute platforms to perform more complex calculations, use larger data sets and get results sooner. In this session, we will explain how vSphere 7, with all of its innovative features, is the perfect platform for data scientists. Being able to run distributed and containerized DL/ML applications natively on vSphere with Tanzu, and fully executed with hardware accelerators with vSphere Bitfusion, shows the true potential of the vSphere platform. This session will also cover design considerations and include demos to showcase how these technologies complement each other.
The Nerdfest VDI Demo: VDI*(AI + ML + Deep learning + GPU) [DWHV2820]
A VDIscover experience session. VDI experts are all about the most innovative technology and pushing limits, including exciting areas like AI, machine learning, deep learning, and the latest and greatest in GPU virtualization. In this session, Dutch VDI expert and consultant Johan van Amersfoort will give a super deep demo, combining Linux VDI with vSphere Bitfusion to run GPU-based deep learning workloads on a virtual desktop. Get a closer look at Bitfusion, the GPU virtualization technology VMware acquired in 2019, and learn how it can apply to data science, developer, and research workloads.
Dynamic Optimization of Mixed VDI and Artificial Intelligence Workloads [ETML1848]
The world has changed. Organizations are being asked to do more with less, and deliver virtual desktops to a remote workforce that can run any application, including 3D professional visualization, and solve critical problems with high-performance computing (HPC) or artificial intelligence (AI). How can organizations maximize their return in both areas while minimizing costs, delivering results and providing an outstanding user experience? In this session, we will show you how to unify the infrastructure for these workloads with dynamic optimization of resources. See how easy it is to turn your NVIDIA vGPU enabled VDI environment into a powerhouse HPC environment, then revert back to VDI as demand changes. We will answer your questions on how to get started with dynamic workload optimization, going beyond VDI by day and compute by night.
How to Architect for Success: Ask the EUC VCDXs [DWHV2826]
A VDIscover experience session. Rolling out a VDI environment requires a million careful decisions, any one of which could have a significant effect on the performance, scalability, cost, and the user experience of your environment. Furthermore, VDI deployments must be designed in the context of the larger EUC environment, considering aspects such as: zero trust security; SSO and conditional access; client devices; web and native apps; and integrating all of these components to make a great experience. Listen as some of the top VDI and EUC experts in the field share their best practices for designing a VDI and EUC environment.
The Future Role and Skills of the VDI Admin: A Panel Discussion (Part 1 of 2) [DWHV2823E]
A VDIscover experience session. Being a VDI admin means becoming an expert on a wide range of technologies, from Windows desktops and networking, all the way to security and cloud platforms. As if that wasn’t enough, VDI and end user computing environments are constantly changing. Where will all these skills take VDI admins next? How do admins home in on what seems like a moving target and decide what new skills to learn? And most importantly, what will the VDI admin role look like in 5 years? In this round table session, top VDI experts explore what the future might hold and how to get ready for it.
The Future Role and Skills of the VDI Admin: Community Discussion (Part 2 of 2) [DWHV3008E]
Being a VDI admin means becoming an expert on a wide range of technologies, from Windows desktops and networking, all the way to security and cloud platforms. As if that wasn’t enough, VDI and end user computing environments are constantly changing. Where will all these skills take VDI admins next? How do admins home in on what seems like a moving target and decide what new skills to learn? And most importantly, what will the VDI admin role look like in 5 years? In this session, four top VDI experts explore what the future might hold and how to get ready for it.
Again, if you would like to schedule one of these sessions, check out this link: https://my.vmworld.com/widget/vmware/vmworld2020/catalog?tab.deliveryfordisplay=1597089771221001YGg0&search=amersf
There are a ton of sessions available to schedule, but there are a couple of them which I look forward to. Some might not be the obvious ones, but still triggered my interest. Here is a top 10 of sessions I really look forward to in random order:
Back-End, Meet Front-End: Modern App Delivery with Tanzu and Workspace ONE [DWMM1468]
As continuous integration/continuous delivery (CI/CD) becomes more ingrained in organizational go-to-market strategies and pushes the boundaries of application development, immense stress is placed on traditional IT infrastructure. VMware Tanzu alleviates this pressure by promoting scale of back-end development, all while VMware Workspace ONE accelerates front-end delivery. Join two End-User Computing consultants and a VMware Tanzu solution engineer as they demo the combined capabilities of each solution to deliver a seamless strategy for modern application development, delivery and monitoring. This will showcase the power of VMware’s portfolio in adopting the CI/CD framework. See how the distinct intersection of VMware Tanzu and Workspace ONE will enable organizations to remove siloes once thought impossible to overcome.
My take: Seeing the Any App strategy in its full force, is something I’ve been waiting for, for a long time!
Solutions Keynote: Journey to a Future Ready Workforce [DW2716]
As remote work takes hold, businesses are seizing the opportunity to evolve their distributed workforce with the tools needed to work in new ways. For its technology, not location, that holds today’s distributed organization together. A fractured foundation means gaps in security, operations and performance, while a digital foundation enables businesses to become leaders in the new work world, delivering seamless workspaces where location is no longer a factor. With Future Ready Workforce solutions, scale access, manage and secure endpoints and optimize the network edge, all through a unified platform where exceptional user experience and intrinsic security are at the forefront. Join us as we’ll bring you the latest insights so you can deploy an agile, complete journey that is easily operationalized for organizational resilience.
My take: A must-see if you are interested in EUC. Basically, Shawn Bass talking and running through demos of all the latest and greatest.
Close Look at Remote: What I Learned from 50 VDI Admin Meetings Since COVID [DWHVDI2827]
A VDIscover experience. Close Look at Remote: What I learned from 50 VDI admin meetings since COVID.
My take: Brian Madden talks you through his experiences with customer meetings. Great to hear about lessons learned from others who manage a VDI.
Deep Dive into Machine Learning Architectures [ETML2517]
Successful machine learning (ML) implementations require a broad and difficult mix of infrastructure, software and integrations—arguably as complicated as the CNCF landscape. In this session, we’ll take a deep dive into a set of field validated architectures for implementing complete ML solutions that enterprise architects and VI admins can deliver today on a VMware platform. We’ll cover scalable data ingestion and storage, massively parallel ML training, subsequent ML model deployment, and day 2 management of ML artifacts using VMware, OSS and third-party software. Key takeaways from this session include field-validated ML solutions, a platform to support your data science and ML engineering teams, and a pathway to scale and manage for growth in ML utilization.
My take: I have been focussing on ML for a while now, as a personal side project. Interested to hear more about this at VMworld.
DevOps for Machine Learning: Enabling Successful ML for Your Organization [ETML2260]
97 percent of IT infrastructure is still on premises, a significant portion of it managed on VMware. Machine learning (ML) is a mandate for enterprise organizations, and running models close to the data source is preferred to reduce cost; increase iteration speed; and satisfy security, compliance and privacy requirements. ML is going to happen in complex digital infrastructures. To support that, I&O leaders need to consider an ML-focused lifecycle that integrates into existing processes. This session will cover: • Key differences between ML and traditional SDLCs • Where the SDLC works with ML, and where it breaks down • An overview of the new ML stack, from training to deployment to production • The five biggest ML infrastructure and process mistakes IT and ML teams commit • Successful case studies and real-world lessons you can use.
My take: One of the things that misses in the data science world, is automation and orchestration and infra as code. Looking forward to this as an extension to my own session about ML and Tanzu (HCP1266).
Game-Changing Employee Productivity with Spatial Computing and VMware [OCTO1213]
Augmented reality (AR) and virtual reality (VR) are two emerging technologies set to significantly improve employee productivity. In this session, you will learn about key use cases and the business benefits of spatial computing, and what an enterprise AR/VR deployment looks like. We will also show AR/VR in action.
My take: Have worked with the ProjectVXR team to test some things in VR. Really interesting to see what they’ve been working on in the last couple of months.
How to Enable Workspace ONE Mobile Flows and Automated Virtual Assistant [DWDE1419]
This session will provide a technical overview of VMware Workspace ONE mobile flows and the automated virtual assistant (AVA) configuration. This advanced-level session will focus on deeper technical aspects, including prerequisites, setup and configuration of Workspace ONE mobile flows and AVA. This session is geared toward end-user computing (EUC) specialists who are familiar with the technology and want a deeper understanding. The assumption for this session is that attendees have previous enablement on the product and hands-on experience.
My take: Flow-based automation is one of the next-gen solutions for organizations from an EUC-perspective. Looking forward learning more about the technical details of VMware‘s solution.
Inside xLabs and Off-Roadmap Innovation in the VMware Office of the CTO [OCTO2406]
The xLabs team in the VMware Office of the CTO incubates new technology solutions with the goal of bringing them to market in anywhere from six months to three years. Attend this session to learn about the range of recently graduated and currently incubating innovation projects crossing a wide range of technologies, including virtual service networks and network slicing, AR/VR, edge computing, cloud native applications, accessibility, machine learning, management and automation, and more.
My take: No explanation needed.
DeepThought: Power Your AI with VMware Tanzu [ETML1895]
Facing challenges in efficient operationalization of your artificial intelligence/machine learning (AI/ML) pipeline? Wondering how to get faster business value from your ML models? In this session, see how DeepThought leverages VMware platforms to manage the end-to-end data science lifecycle, providing scalable compute and storage via one-click Kubernetes deployment with VMware Tanzu for on-premises and hybrid cloud. The solution includes a fully automated DevOps pipeline with preconfigured GPU and CPU images, data integration, and distributed model training. Integrations with Slack, Teams and SSO streamline operations, while VMware Tanzu Observability by Wavefront offers model governance, performance monitoring, data drift and concept drift. This session includes an overview and technical walk-though of DeepThought, followed by a demo.
My take: I have integrated Tanzu and Bitfusion in my own lab and used it for a personal project (STIG). Interesting to see how others are using this.
What’s New with DRS and vMotion in vSphere 7 [HCP1246]
With VMware vSphere 7, we introduced a completely new, more workload-centric vSphere Distributed Resource Scheduler (DRS) logic. DRS now provides better insights on why VMs are happy or not. In this session, learn how the DRS algorithm was updated to support next-gen workloads, including vSphere with Tanzu. Find out how DRS can now apply workload placement logic when workloads (such as machine learning/artificial intelligence) use hardware accelerator devices (such as GPUs or FPGAs). vSphere vMotion was also heavily updated to ensure that large workloads (such as transactional database platforms) in both VMware Cloud and on-premises vSphere can be live-migrated far more efficiently and effectively by reducing the migration and stun time involved. See how customers can be comfortable using vMotion for large (or monster VM) workloads.
My take: Always fun to see my buddy Niels present. In this case, some new improvements to vSphere that will increase the resiliency of data science and ML apps.
There’s just too much to choose from. Luckily enough, every session will be available on demand when the event is over. With 1000+ sessions, I guess that will cover my own learning demand until VMworld 2021 🙂
- VMworld 2020 – My take and my TOP 10 sessions - September 4, 2020
- Announcing vSphere Bitfusion as GPU acceleration platform - June 2, 2020
- Mastering voodoo: Building Linux VDI for Deep Learning workloads, Part 2 - April 2, 2020