Moving Beyond CAD and BIM

In the construction sector, it’s not just users of Computer Aided Design (CAD) or Building Information Modelling (BIM) software that requires access to GPU-accelerated workstations. As rich 3D models start to be re-used and data flows from design all the way through to construction, it’s also extended architecture, engineering and construction (AEC) teams that need GPU-accelerated machines. This could be for design/review, 4D construction planning or design visualization – or simply just to aid communication.

The challenge for AEC firms is how to provide teams with appropriate hardware. Viewing a complex 3D BIM model on a PC, laptop or a standard Virtual Machine (VM) without GPU acceleration can result in a very poor experience. Basic view operations such as pan, zoom and rotate may become choppy and it can be very hard to position the model on screen. In some cases, the 3D application simply becomes unusable.

GPU-accelerated workstations are specifically designed to handle complex 3D datasets, but AEC firms can find it hard to justify their cost for wider teams, especially when the powerful 3D graphics capabilities might only be needed on occasion throughout the working week.

A GPU-accelerated virtual desktop, delivered via the Cloud through AMD powered Microsoft Azure NVv4, offers the perfect solution for the part time consumer of 3D models – be that a project engineer, senior architect or site manager.

Power of the Cloud

Delivering GPU-accelerated virtual desktops via the Cloud offers additional benefits. Providing there is a high-bandwidth, low latency Internet connection in place, 3D applications can be accessed from anywhere on any supported device. With Microsoft Azure NVv4, you can stream 3D applications to a tablet on a construction site, or a home PC to support flexible working – even a laptop in a different country.

The other benefit of VMs in the Cloud is that applications run on Microsoft Azure servers, right next to your data, and only encrypted pixels are streamed to the end device. It means users can get instant access to up-to-date project data and do not have to wait for huge BIM models to download or sync. Storing all project data in the Cloud and not on users’ personal workstations also offers big benefits for collaboration, security and revision control.

Complete flexibility

In the same way that a personal workstation can be configured with different CPUs, GPUs, memory and storage, Microsoft Azure offers a range of VMs to suit different workflows. These are known as “instances.”

The beauty of the new GPU-accelerated Microsoft Azure NVv4 instances is that they are designed specifically to cater to a variety of 3D workflows. The instances are powered by AMD Radeon Instinct™ GPUs.

Previous GPU-accelerated Microsoft Azure VMs could only be configured with a full GPU, but this meant customers could end up paying for resources they did not need. While most BIM-centric workflows require some form of 3D graphics acceleration, many do not need such high levels of 3D performance.

By using AMD Radeon Instinct™ MI25 GPUs, Microsoft Azure NVv4 instances offer the flexibility to cater to a wide range of GPU-accelerated workflows. Customers can choose from four different VMs, including one with 1/8th of a GPU, which might be suitable for viewing a simple BIM model, all the way up to one with a full GPU that could handle a rich 3D model for design visualization.

Matching instances more closely to workflows can make GPU-accelerated virtual desktops much more cost-effective than ever before. As AMD powered NVv4 instances are billed by the hour, customers can also scale resources up and down to support the ever-changing needs of projects over time, both in terms of performance and the number of VMs required by the project teams.

Three workflows where AEC firms can benefit from Microsoft Azure NVv4

1. Design and review – NVv4 enable better collaboration across departments and staff

Design / review software comes in many forms, from simple view, measure and mark-up, to model co-ordination and clash detection. In addition to using BIM models it can include large reality meshes or point clouds.

Design / review can be done in a single room with one team gathered around a large screen. Alternatively, teams can collaborate on the same 3D model from multiple locations, even on a construction site. With Microsoft Azure, as everything is hosted in the Cloud, it’s ideal for this distributed approach. In addition, as design / review might just happen once a week it’s also well suited to the flexible, per hour charging model of AMD-powered NVv4.

Design / review is not just a static process. Once any interferences, clashes or general issues have been identified, they need to be resolved. Issue tracking then allows teams to share tasks and assign work responsibilities, which means individuals may need further access to GPU-accelerated VMs.

2. 4D construction planning – NVv4 delivers flexibility and access to your information

Planning for construction projects has historically been managed through Gantt charts. However, there are now an increasing number of software tools that can be used to create 4D simulations, linking 3D BIM models to the project schedule.

Project managers can use 4D software to plan construction sequences from beginning to end. Microsoft Azure NVv4 instances make it possible to work fluidly with simple or complex models, zoom into details or play entire animations of the construction process.

4D software is not only used by the project manager. The 4D model can be used for collaborative planning and decision-making, and software that simply views 4D construction models is freely available for multiple stakeholders.

3. Design visualization – NVv4 gives designers the performance they need, while on the move

With an increasing number of easy-to-use real time viz tools, design visualization is being used much more frequently at all stages of architecture, engineering and construction for better understanding and communication.

The resulting viz assets are not just for the eyes of architects or viz specialists. Extended teams, especially nontechnical users, as well as clients or members of the public, can also benefit greatly. Many viz tools can now push content to a browser or produce an executable for distribution.

The GPU requirements to view these assets can be quite high. Laptops or mobile workstations with powerful GPUs are not only expensive and thus hard to justify for infrequent use but are also large and not very portable. Microsoft Azure NVv4 gives extended teams the power on demand to view these visually rich assets. Instances can be matched to the complexity of the model.

AMD technology changing the rules for GPU-accelerated VMs

The introduction of AMD-powered NVv4 instances is shifting the expectations for VM deployments and is sure to have IT managers taking note. What’s new? Well, The NVv4 instance is the first VM on Microsoft Azure to take advantage of SR-IOV technology (single-root input/output virtualization), which supports GPU partitioning through AMD MxGPU GPU sharing technology, which is built into the AMD Radeon Instinct MI25 GPU.

For customers, this means greater flexibility, enabling the entire enterprise – and not just the CAD users – to enjoy dedicated GPU-accelerated Virtual Machines (VMs), delivering an exceptional application experience regardless of the workloads.

Customers have four distinct NVv4 instance options to choose from, scaled to share a single GPU’s resources among as many as eight VMs. Alternatively, IT managers can maximize the user density of NVv4 with Windows® 10 multi-sessions, supported by Windows Virtual Desktop with available plug-ins from Citrix® and Teradici®.

SR-IOV technology plays a very important role when it comes to security. It allows the isolation of PCI Express® resources between different users. It is already the standard used to share networking resources (NICs) and secure network traffic. Each resource has Virtual Functions (VF) associated and each VM can only access the physical resource via its own allocated VF. This helps ensure that each VM is isolated from others e.g. memory is secured and not shared. NVv4 Virtual Machines will initially be available in the South Central US and West Europe Azure regions and will be available in additional regions soon thereafter.