This week I’m at NVIDIA’s GTC conference in my new home town (since December 2012) of San Jose, California, and having a great time. I’ve seen some fascinating sessions but most of all it’s been a great chance to meet up with a lot of both new and old customers attracted to Citrix and to virtualization by our great 3D graphics support. If you are lucky enough to be at GTC, please join me on Thursday at 2pm to see how established technologies such as NVIDIA GRID vGPU (introduced with XenServer 6.2 SP1 last year), Citrix XenDesktop and XenApp GPU sharing and dedicated CUDA CPU usage are being used today by customers in Healthcare, Aerospace, Manufacturing Engineering and Architecture, using AEC, CAD, CAE and PLM/PDM software to enable their core business. I’ll be talking through some case studies of real customers undertaking projects including:

  • Building F16 fighter jets
  • High-resolution FDA-regulated cancer diagnosis and treatment
  • A major architectural firm constructing some of the world’s most beautiful skyscrapers
  • A production vGPU university cloud enabling students to work anywhere, cutting costs and taking advantage of Citrix’s favorable educational licensing
  • An engineering company who sped up their Ansys FEA CAE simulations by 400% via virtualization

GTC has also been fantastic for me to touch base with a lot of our partners and their customers with whom I’ve been working on joint projects since XenDesktop HDX 3D Pro was introduced in 2009. It’s fascinating hearing how customers have been using the new reference architectures we’ve been working on for GRID vGPU and our other 3D graphics technologies with Dell (see here),  HP, IBM and others, and it’s definitely worth checking out the server vendors’ demo booths!

At the moment I’m putting the finishing touches to my own talk for GTC, scheduled for Thursday at 2pm. At a conference that sees a lot of announcements and previews about technologies yet to come, I’m taking a different approach this year. In the past we previewed GRID vGPU and my team spent a lot of time last year working with our partners and customers hardening the Citrix offering, certifying and benchmarking CAD software, stress testing GPU servers and designing large scale reference architectures. Some of the customer production systems I’ll be discussing are now several years old and we’ve used our experience to feed into incremental development. And high performance hardware-based GPU sharing is now successfully in production at companies around the world, leveraging the multi-user capabilities of NVIDIA’s latest GRID GPUs. So I’m planning to delve into some of the nitty-gritty details of how these technologies actually work in the field and cover themes such as:

  • Experiences with specific software packages that have been virtualized and certified such as Autodesk AutoCAD and Revit, Siemens NX and TeamCenter, PTC Creo,  Microsoft Office,  Adobe Creative Cloud and others, along with a few insider tips on optimization…
  • Cost effective solutions for “second tier” users
  • Integrating GPU accelerated technologies with networking and products such as NetScaler to mitigate network latency and provide secure access
  • Specific usage of GPU enabled servers such as the Dell R720 and HP DL380p, and the use of Macs and iPads as clients
  • Private cloud usage to control proprietary design data and confidential personal medical data

 As Director of Product Management for HDX, I oversee a broad, integrated portfolio of technologies associated with user experience and interaction that extends beyond GPU technologies. Many of these have been essential for engineers and designers to leverage 3D graphics in the type of real customer scenarios I’ll be discussing, and I’ll be happy to chat with anyone who wants to talk about integrating GPU technologies in virtualized systems with smart cards, 3D mice, printers, thin clients, technologies to reduce WiFi jitter, optimized and lossless compression, etc.

Here’s the full abstract for my talk:

S4581 – Customer Experiences with GPU Virtualization and 3D Remoting

Derek Thorslund (Director of Product Management, Citrix)   Highly-Rated Speaker

  • Industry leaders have been delivering centralized 3D graphics applications using Citrix XenDesktop HDX 3D Pro since 2009, and now the cost per user is more attractive than ever thanks to GPU-sharing technologies including XenServer/NVIDIA GRID vGPU. Learn about real-world customer experiences with GPU virtualization and 3D graphics remoting, including use cases, benefits and business drivers, scalability measurements, bandwidth consumption per session, cost saving, and best practices.

Session Level: All

Derek Thorslund
Director of Product Management, HDX