Wednesday, August 14, 2013

Performance on-demand with cloud computing

In previous HPC blog posts, we’ve discussed several methods for speeding up your simulations with high-performance hardware. However, this hardware can be a significant investment for a company or research group.

For example, the costs of a cluster for MPI computing are not limited to purchasing the hardware – it includes enterprise-level interconnects and, if necessary, GPU cards. There is also the cost of installation to consider, along with the need for a dedicated technician whose job it is to maintain the system, training for users and the rapid obsolescence of HPC equipment. For customers with very large workloads, an MPI cluster can be a worthwhile investment, but for smaller companies, occasional simulations probably won’t justify this expense.

For these users, cloud computing provides a very useful connection to the world of HPC. Instead of using computing resources on-site, they can upload their models to a cluster owned by a cloud computing provider such as Bull extreme factory. They then hire time on the cluster to carry out the calculations, and when they are complete, view the results with a visualization session or download them.

Since you only pay for the time when you’re using the cluster, cloud computing can be very economical for occasional jobs. To help these users, we offer special cloud computing licenses with durations from as little as a week. Because CST STUDIO SUITE® comes pre-installed on the cluster, you can begin using it right away.

The cloud computing workflow is simple, and doesn’t differ greatly from the usual workflow that our users are familiar with. Our portal on Bull extreme factory allows you to manage your data, licenses and simulation tasks through a simple webpage, and it also lets you start a visualization session. This launches a window with the CST STUDIO SUITE user interface, allowing you to control and view the simulation just as it would appear if it was running on your own workstation. Settings can be adjusted easily, and results can be viewed without having to download the entire file.

Cloud computing is very flexible – you can choose how powerful a cluster needs to be for the job and when  you need it – and while very large companies, who need HPC almost constantly, might find that owning a dedicated cluster is still the most efficient option, cloud computing levels the playing field for small and medium companies. Security is of course a key concern for many companies, and so data can be transferred using a secure channel with HTTPS and, where available, a VPN. Additional security features are available on demand.

Cloud computing clusters are well-equipped to take advantage of the HPC features of CST STUDIO SUITE, with multi-core processors, GPU cards and high-speed interconnects. These features can be activated using acceleration tokens in the same way that local resources are.

With the introduction of cloud computing, CST STUDIO SUITE now has a HPC solution for almost every need and price range, from entry-level to enterprise. See the HPC section of our website or contact your local sales representative to discuss the best HPC option for your requirements.

Thursday, August 8, 2013

Pushing the limits of simulation with MPI computing

While there are many ways to improve the performance of a workstation, there are always limits on what a single computer can do. When faced with a problem that would be impractical or impossible to solve on one computer, MPI computing offers a way forward.

MPI, which stands for Message Passing Interface, is a system for allowing a computer cluster to act like a single supercomputer. In a simulation with MPI computing, the model is broken down into multiple simulation domains, and each computer in the cluster is sent one of these parts to solve. Unlike distributed computing, these calculations are not independent – after all, electromagnetic waves can pass freely from one domain to another. Domain decomposition is possible because the nodes of the cluster regularly exchange field data at simulation domains, so that the total field across the device can be calculated.

In CST STUDIO SUITE®, MPI computing is currently supported by the transient solver, frequency domain solver, integral equation solver and wakefield solver. While these solvers are powerful, the use of one computer limits the size of the models that can be. With MPI computing, the model is broken down into smaller domains, removing these restrictions – if the cluster is large enough, models of arbitrary size can be simulated. When using the transient solver, MPI computing can even be combined with GPU computing; a cluster with multiple GPU cards overcomes the usual memory limits of a single GPU.

Customers sometimes ask whether they can use MPI computing on their ad-hoc clusters, using workstations or small enterprise servers connected by a standard Ethernet. Unfortunately, this is not really practical – to use MPI computing effectively, it needs to be run on a dedicated supercomputer-type cluster with homogenous nodes and a high-speed interconnect like InfiniBand®. The nodes  have to exchange very large amounts of data between each other constantly, and a slow connection will completely negate the benefits of MPI computing. An alternative for these users is to use cloud computing.

The final blog post in this series will explore cloud computing for HPC, and show how it makes the power of MPI computing available to users who don’t have the resources for a dedicated cluster.

Thursday, August 1, 2013

Boosting simulation performance with GPU computing

GPUs (graphical processing units) were originally designed for generating video graphics, but their high memory bandwidth and the ability to perform calculations in parallel very quickly made them attractive to scientific and engineering users. GPU cards typically have hundreds of parallel cores, along with several gigabytes of on-board RAM. The memory bandwidth for GPUs is much higher – a Tesla K20 has a bandwidth of 208 GB/s, while the widely-used DDR3-1333 RAM in a quad-channel configuration offers just 51.2 GB/s – and this leads to significant speed advantages in certain situations.

CST STUDIO SUITE® supports GPU acceleration with the Nvidia® Tesla series of GPUs for the transient solver, the integral equation solver and the PIC solver. Multiple GPU cards can be used together for even greater speed-ups – either combined in one machine or distributed across a cluster – although PIC solver currently only supports one GPU per simulation. As part of an oPAC research network project, we’re currently developing a multi-GPU PIC solver, so watch this space!

One advantage of GPU computing is that it is very scalable. GPU cards are available for a wide range of computer types, from individual workstations up to servers for computing clusters. Distributed computing (DC) can both take advantage of GPU computing, as can MPI computing with the transient solver. The acceleration tokens in CST STUDIO SUITE can be used to choose the best combination of HPC techniques for each simulation. For further information about acceleration tokens, see the licensing guide.

The next blog post in this series will discuss MPI computing in more detail, and show how it can be used to simulate problems that are beyond the capabilities of a single computer.