Reduce Manufacturing Costs with HPC

Print Friendly, PDF & Email

Last week we looked at hurdles facing the manufacturing industry. This week we are focused on another major consideration for today’s manufacturers which is the the pressure to reduce manufacturing costs.  This, in turn, is driving the need to reduce expensive physical prototyping through the use of modeling and simulation earlier in the design process.

SGI MFG WP - Cover Sm.This article is part of 5 article series on how HPC is helping to Fuel the Growth in Manufacturing. This white paper was created by the editors of insideHPC and underwritten by SGI and Intel.

Intense competition on a global basis requires manufacturers to speed their time-to-market, which can only be accomplished by a more efficient IT infrastructure and a major boost in end user productivity.

Competitive advantage can also be gained by using HPC and CAE to address the complex physics inherent in the design of today’s sophisticated products.

Meeting these challenges is far from easy.  For example, computing larger models to generate more detailed simulations of both the product and the processes used on the factory floor, can take days, even weeks, to run on typical workstations or HPC cluster. And CAE workstations, the solution of choice for most small- to medium-sized manufacturers, are expensive and require costly and complex software licenses from the various ISVs.

Because of the large datasets involved, sharing data among engineers is difficult.  Also, as the computational capabilities of these users grow with access to ever more powerful workstations and software, the IT department is hard pressed to keep up with their demands.  The fact is, in the world of manufacturing, HPC usually is not a core IT competency; the hiring and retention of experts in such disciplines as CFD and FEA modeling, simulation and analyses, is not a line item on most small to medium-sized manufacturer’s budgets.

Meeting the Challenge

In an ideal world, a CAE system that can address these challenges would not be a box of components and wires dropped off by the vendor that turns into a six month science project.  Instead, the ideal system in an ideal world would be:

  • Flexibly architected
  • Resilient
  • High performance
  • Cost-effective
  • Easy to deploy
  • Easy to administer

With these goals in mind, SGI has created a solution that simplifies high performance computing for CAE.  SGI has created an easy to deploy, unified platform that addresses a broad spectrum of CAE applications and workloads.  It allows you to leverage your investments by simplifying integration with existing infrastructures and maximizing the efficiency of available licenses and personnel.  The system optimizes your productivity by providing faster data access to do real work and increases the value of archived data.

Next week we take an indepth look at HPC computing and storage capabilities required to handle typical CAE workflows.  If you prefer you can download the complete article series from the insideHPC White Paper Library courtesy of SGI and Intel.