This special report sponsored by QCT discusses how the company can work with leading research and commercial organizations to lower the Total Cost of Ownership by supplying highly tuned applications that are optimized to work on leading-edge infrastructure. By reducing the time to get to a solution, more applications can be executed, or higher resolutions can be used on the same hardware. QCT also has experts that understand in detail various HPC workloads and can deliver turnkey systems that are ready to use. For customers that wish to modify source code or that develop their own applications, QCT supplied highly tuned libraries and extensive guidance on how to get the most out of your infrastructure, that not only includes servers, but networking and storage as well.
This technology guide, insideHPC Special Report Optimize Your WRF Applications, shows how to get your results faster by partnering with QCT.
Introduction
Investment in High-Performance Computing (HPC) needs to show a return on investment from the start. Merely buying a bunch of servers and expecting your applications to run faster most likely won’t work. Understanding the underlying hardware and how applications can take advantage of the critical computing power is one of the most critical aspects of using HPC. By partnering with a supplier that has the domain expertise and the technical know-how to work with various organizations will pay off in the long run.
Servers and More
A modern computer server contains several necessary components that need to work seamlessly together to deliver fast results. A necessary HPC infrastructure will include many servers, network interconnects, and storage that is suited to the capacities needed for the applications that will be executed.
Typical Server Resources:
- 1,2 or 4 sockets
- 16+ cores per socket
- DRAM Memory – 512 GB or more
- Internal storage
- Networking capability
While many companies today design and manufacture servers that contain the above hardware, many of these companies do not have the expertise to design a full solution that meets demanding customers’ performance requirements and cost constraints. Domain expertise is critically important when working with a server vendor. This expertise leads to a more optimized hardware and software architecture for the given workloads that a customer must execute. By understanding in great depth these workloads, a valued partner can experiment with sophisticated technologies to determine the ideal combination of software and hardware. This leads to impressive gains in the performance of the overall system, which can contain many thousands of servers or computing units. It is especially important to understand the interaction of the software and the underlying hardware architecture, at a level that many end-users will not understand. Development environments such as the compiler used, the libraries used, and the communication choices will all affect the performance of critical workloads.
The knowledge that a server and storage supplier bring to specific industry domains cannot be overlooked for several reasons, including:
- Proper sizing of infrastructure – Customer receives solution based on requirements
- Expert advice on server components – Customer gets the best matching of CPUs, Memory, Storage
- Tuned software stack – Customer can get productive faster
- Knowledgeable Sales Engineers – Customer can focus on science rather than optimizing the system
Understanding Climate Change
One of the most pressing, if not the most severe challenge of our time is to understand the consequences of climate change. Research is being performed around the world to determine the extent of damage, both physically and politically, that will are a result of climate change. Also, research into mitigation options and their effect on the climate continues to be an active topic.
The time to solution will typically decrease in time as faster processors become available. Still, there is also room for improvement in the programming of the physics involved and making better use of the underlying hardware infrastructure.
A popular application that simulates climate change is the Weather and Research Forecasting (WRF) model, which is a collaborative partnership of the National Center for Atmospheric Research (NCAR), the National Oceanic and Atmospheric Administration (represented by the National Centers for Environmental Prediction (NCEP) and the Earth System Research Laboratory), the U.S. Air Force, the Naval Research Laboratory, the University of Oklahoma, and the Federal Aviation Administration (FAA). WRF is used by thousands of researchers worldwide in over 150 countries.
Many scientific application codes are written in FORTRAN, due to several reasons. These include:
• Easy to learn
• Optimized numerical libraries
• Extensive collection of supporting applications
Since WRF is open-sourced, organizations can download the application code, make changes or add features and compile and deploy on their on-premises servers or work with others to use their new version. The choice of compilers for the source code can have a significant effect on the ultimate performance of the application. Customers who download the source code typically want to see results and are not interested in testing many different compilers with many different flags set. There are many compilers available for FORTRAN today, and many end users look for guidance as to which one of these compilers will result in the highest performance. The goal of the end-user is to obtain answers as quickly as possible and not experiment with different compilers or compiler options.
Over the next few weeks we will explore these topics surrounding WRF applications and how you can get your results faster by partnering with QCT:
- Introduction, Servers and More, Understanding Climate Change
- Introduction to WRF
- Benchmark Results
Download the complete insideHPC Special Report Optimize Your WRF Applications, courtesy of QCT.
