SPONSORED CONTENT HIGH PERFORMANCE COMPUTING


not be able to maintain the system or to deliver the required performance from their applications. For businesses that do not have the


internal infrastructure and staff to manage a cluster, or startups and small firms that want to bypass that investment, or want to adopt a hybrid approach to managing compute resources, pre-configured HPC solutions tailored to simulation can combine performance and usability. Compute-intensive simulations found in


today’s industrial design projects require computational resources that can deliver modelling and simulation workloads at scale. To support those applications, modern HPC systems require multi-core processors, high-bandwidth fabrics, fast storage and sufficient I/O capabilities to not bottleneck the performance of a given application. To deliver this kind of hardware,


typically an organisation would need staff or additional help to build a balanced architecture to support the specific set of applications.


While this is a complex process, there is a risk of buying sub-optimal hardware, then wasting a potentially expensive component that cannot be fully utilised because of a memory or I/O bottleneck


“While buying a general purpose solution may deliver the performance an organisation requires, it may not be the best solution for users with specific CAE application requirements”


for example. The complexity and variety of technologies available on the market mean that designing and assembling an HPC system for specific workloads can be time-consuming. This is particularly true if there are a variety of applications that need to be benchmarked or validated with a new architecture. Pre-configured solutions, such as the


Intel Select Solution, reduce the burden of designing, provisioning a cluster that must then be optimised and maintained to deliver consistent service for the users. Employing these technologies provides the peace of mind that your application will run effectively and deliver the performance required without large-scale optimisation, because it is designed to run


ISV codes commonly used in simulation today.


Whether an organisation wants to


deploy cloud or in-house resources, build its own cluster or buy a pre-configured solution, the total cost of ownership must be considered.


While buying a general purpose


solution may deliver the performance an organisation requires, it may not be the best solution for users with specific CAE application requirements. Efficiency, infrastructure costs, staffing and electricity bills can all drive decision making to a reduction in the total expenditure on computing infrastructure. To deliver the most efficient solution possible, organisations can look to deploy a cluster which has been specifically built and optimised for CAE applications. These systems can be tested ahead of time, which means that engineers can understand how fast their real-world workloads will run on a specific hardware configuration.


References 1.


2.


www.grandviewresearch.com/industry-analysis/ computer-aided-engineering-cae-market


www.intel.com/content/www/us/en/products/solutions/ select-solutions/hpc/simulation-modeling.html


New White paper now online


VIEW FOR FREE*


Overcoming challenges of HPC


Clusters for Simulation & Modeling (Boston)


Engineers working in an increasingly broad range of industries are using HPC to execute simulation workloads. The Boston Intel Select Solution enables organizations to scale quickly with a workload-optimized solution that contains the latest HPC hardware to help organizations bypass the challenges associated with setting up a HPC cluster.


www.scientific-computing.com/white-papers


SCIENTIFIC COMPUTING WORLD


www.scientific-computing.com | @scwmagazine Autumn 2020 Scientific Computing World 15


*Registration required


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38