Update: NVIDIA have released the webinar recording: CIMdata Presentation: Secure Distributed Visualization for High Performance 3D Data Users; I’d still recommend reading the whol whitepaper it was based on though – fascinating!
There’s been a lot of great reviews and feedback on the Citrix-NVIDIA vGPU and GPU pass-through technologies applied to CAD. CAD is pretty and illustrates the technologies well. PLM (Product Lifecycle Management) and PDM can often get overlooked because well a shiny simulation of a car does look great.
I spent nearly a decade working for Siemens PLM and learned never to overlook the importance of the hidden PLM workhorse nor to underestimate their value. When we first previewed the vGPU technologies, credible CAD industry analysts like Evan Yares started highlighting how a technology like vGPU could help PLM specifically.
So what caught my eye about this particular webinar? The caliber of the presenters and the fact it was driven by data from PLM analysts CIMdata. CIMdata have been researching PLM virtualization for a long while now and the data was presented by Stan Przybylinski who spent many years working for Dassault’s Enovia; when it comes to PLM you don’t get much bigger or experienced.
The webinar centered around a free whitepaper available from CIMdata “Virtually Delivered High Performance 3D Graphics: Secure collaboration with 3D data on all devices”.
What I liked was that this wasn’t about the presentation of a virtualization technology but based around real research into where, how and what PLM users were actually doing and wanted to do. In reality many of the designers will still be in the research and design departments but it is the PLM users who stand most to gain from technologies like vGPU. It’s down on the factory floor or out at the customer site where remote access to the PLM data and supply chain is needed.
CIMdata’s report looked at the changing pattern of remote device access amongst PLM users. They also looked at how PLM systems involve a large number of third party applications and suppliers and the need to work collaboratively whilst protecting your own data in a secure datacentre, which is what virtualisation technologies like XenServer are all about.
The CIMdata is a fascinating read looking at how PLM workers work cross-geographies, the breakdown of device use, the cost-reductions virtualisation offers when upgrading hardware and software as well as how virtualisation can deliver disaster recovery and data protection. I found their investigations into changing patterns of outsourcing and data access particularly insightful.
vGPU makes accessing PLM on a mobile device viable in terms of usability but also cost-effective. With the vGPU technologies initially scaling to 64 users per card we are finding that actually because most PLM users need GPU enablement and access for part of the day and maybe in shifted time zones it is possible for PLM applications to scale a very large number of users with a relatively low hardware investment.
Also joining the panel was Jonathan Goldstine from IBM, with decades of experience in mainframe virtualization Jonathan and his team were working in large scale virtualization before Citrix was founded. IBM has been responsible for a large number of the world’s most complex and enormous PLM deployments and with solutions experts with such experience of the challenges and rigour of large scale tier 1 automotive and aerospace I’m very excited about the partnerships being built up around the XenServer and NVIDIA vGPU technology. IBM are investing in the IBM iDataPlex dx360 M4 and IBM Flex System as certified platforms for NVIDIA GPU technologies.
vGPU has a lot to offer the big PLM brands such as Siemens PLM TeamCenter, SAP’s PLM offerings and Dassault’s Enovia. With PLM getting more sophisticate with products and components involving more search and lightweight editing such as Geolus and ShapeSpace.
The related supply chain products supplying parts are applications whose uptake of virtualisation, cloud computing and GPU usage I’ll be watching with interest, products and services such as Cadenas and GrabCAD. As are the material information services available to PLM, products such as Granta Design’s MI and Key To Materials available on the Altair Hyperworks platform
I wrote a blog touching on why vGPU works with Citrix’s other technologies to deliver CAD/CAE and PLM solutions a few weeks ago. Really though it is a technology that you should try for yourself though so please do visit our vGPU tech preview page. You will just need a valid MyCitrix login.
Update: If you found this post of interest please come to our webinar on 11th December 2013, register here.