Learn how Arm-based systems from HPE, powered by Fujitsu CPUs, make new HPC technologies accessible to O&G applications—and unlock high performance and memory bandwidth for the most complex codes.
By Scott Duplantis, HPE Chief Technologist – Oil and Gas
Oil and gas producers are investing in HPC to optimize their most demanding applications. But HPC adoption is more than just “deploy and go.” Companies that implement GPU accelerators for HPC often struggle with the complexity and cost of rewriting code that was not originally developed for accelerators.
Unless new solutions are compatible with existing workflows and applications, O&G companies have to refactor large volumes of existing source code to run on the specialized hardware. The process of refactoring or rewriting in-house code can be extremely labor-intensive, experimental, and error-prone.
For many organizations, it is more productive to invest their software engineering effort in advancing the capabilities of their applications. After all, this is where their IP is encoded, and competitive advantages are developed from simulations.
If they can achieve similar performance using CPU-based computation, they can realize the benefits and strengthen their capabilities very quickly. O&G producers can achieve the right compute using a new breed of CPU that delivers world-leading performance. These breakthrough systems create a robust HPC environment with scalable compute and high memory bandwidth to increase ROI for existing applications. This presents an extraordinary opportunity for O&G companies to modernize without having to refactor, so they can focus on pursuing growth and value.
HPC innovations for the future of O&G
Today’s energy producers are working to meet the demands of compute-intensive processes with a purpose-built CPU that is accessible to existing O&G applications. The Fujitsu A64FX is an Arm-based processor in the HPE Apollo 80 system that makes this possible, delivering a technology infrastructure with high memory-CPU bandwidth and an insatiable appetite for geo data.
By rebalancing the architecture using this powerful CPU with directly attached high bandwidth memory, the powerful A64FX CPU receives data at extraordinary rates and is not left waiting for data to compute. This memory-CPU bandwidth matched with arm scalable vector extensions (SVE) enables performance increases without requiring accelerators. The user can allocate precious time and talent to improving capabilities and tuning for performance, rather than rewriting their code.
These supercomputing class systems are bringing the power of HPC to memory-bound, floating point intensive workflows and empower engineering leaders within O&G to drive major improvements:
- High throughput and scalable memory bandwidth with CPU capacity
- Streamlined coding and porting of applications
- Rapid path for deploying and tuning applications
- Increased rate of innovation driving application performance
If your company is facing refactoring, look no further. HPE is empowering O&G companies to enhance their technology environments without the repercussions of refactoring.
Arm-based systems from HPE, powered by leading Fujitsu A64FX processors, bring the benefits of GPUs to the CPU without the labor and costs of making major changes to application code. Porting codes to Arm-based systems is relatively simple, so startup and development expenses are low, and the ROI begins quickly.
HPE Apollo 80 Arm-based systems powered by the Fujitsu A64FX™ bring new technologies to HPC for the first time, including SVE and direct-attached high bandwidth memory to unlock high levels of performance for HPC applications and complex codes. SVE was designed specifically for HPC software codes and can vectorize elaborate structures, such as gather-load, scatter-store, and horizontal reductions and allows auto-vectorization of codes.
To help with developing, testing, and optimizing software, HPE offers a comprehensive software stack for HPE Apollo 80 Arm-based systems that includes software for system administration, software development tools for HPC applications, and workload management tools to run jobs efficiently.
HPE Cray Programming Environment is a fully integrated software suite with compilers, tools, and libraries designed to maximize programmer productivity, application scalability, and performance. These capabilities are essential for O&G companies that develop their own HPC code and port to the new SVE architecture—the same developer tools used to support multiple target platforms.
By simplifying porting of existing O&G applications with minimal recoding and changes to the existing programming models, companies can ease the transition to new hardware architectures and configurations. With extensive experience and deep expertise from HPE and Fujitsu, SVE will be a standard for Arm-based HPC systems that will help make applications future-ready.
HPC application performance is critical for upstream apps like seismic interpretation and reservoir modeling, mid and downstream applications like fluid dynamics, and emerging predictive applications that are improving operations, analyzing risk, and developing new core competencies critical for “next era” success.
HPC has the potential to drive improvements throughout the O&G value chain—and we can help. Contact us today to learn more.
Scott is Chief Technologist for oil and gas top accounts at HPE. He leads the vertical initiatives to develop industry-aligned messaging and solutions in support of sales teams and the customer base. Scott’s background is in IT technology consulting and sales and IT technology operations.
Advantage EX Experts
Hewlett Packard Enterprise
twitter.com/hpe_hpc
linkedin.com/showcase/hpe-ai/
hpe.com/info/hpc
AdvEXperts
Our team of Hewlett Packard Enterprise Advantage EX experts helps you dive deep into high performance computing and supercomputing topics.