Challenging the barriers to High Performance Computing in the Cloud

By Bala Thekkedath, Global HPC Marketing Lead, AWS

January 7, 2020

Cloud computing helps democratize High Performance Computing by placing powerful computational capabilities in the hands of more researchers, engineers, and organizations who may lack access to sufficient on-premises infrastructure. The cloud’s flexibility and scalability offer virtually unlimited capacity, eliminating wait times and long job queues. Access to new and evolving services and applications make it easy to evolve and modernize workflows, like incorporating Artificial Intelligence (AI) with HPC. With HPC in the cloud, organizations only pay for the capacity they use, and there’s no risk of on-premises infrastructure becoming obsolete or poorly utilized. In addition, cloud-based services enable innovation without constraints by delivering faster results and improved flexibility. AWS gives organizations the power to create HPC clusters on demand, instead of waiting for equipment to be built—helping drive business insights and organization productivity.

Despite its advantages, some organizations remain hesitant to move their HPC workloads to the cloud, due to questions about cost, security and performance. With today’s clouds, these assertions are outdated and generally inaccurate. HPC on AWS, powered by Intel® Xeon® Scalable processors, offers the most elastic, scalable cloud infrastructure to run HPC applications, and the range of services makes it easier than ever to get started quickly, securely, and cost-effectively.

 

Cost and Cost Management

For many organizations, the cost of running HPC in the cloud is a major concern. In a recent market survey conducted by a third party for AWS, almost half (49%) of participants said cost and cost-management were barriers. When considering the cost of cloud-based HPC systems, organizations should note that in many cases a basic TCO analysis often does not tell the whole story. Demand for on-premises HPC resources often exceeds capacity and lost productivity due to an over utilized system has massive implications for organizations that place high value on the pace of innovation. Moving HPC workloads to the cloud also eliminate the need for periodic technology and infrastructure refresh cycles every three to five years, ensuring that innovation continues at a rapid pace

AWS delivers an integrated suite of services that provides everything needed to build and manage HPC clusters in the cloud, simply and cost-effectively. There are no upfront capital expenditures or lengthy procurement cycles, and the only cost is for capacity used. It offers flexible pricing models that provide significant cost savings for time-flexible, stateless workloads. AWS constantly delivers new services and features, like 2nd generation Intel Xeon Scalable processors with Intel® Deep Learning Boost (Intel DL Boost) to enable new capabilities, improved performance, and optimization for all current HPC frameworks. AWS offers cost management and analysis tools such as AWS Cost Explorer and AWS Budgets. Additionally, AWS partners like Ronin have built cost-control models on the platform.

 

Data Security and Data Governance

Concerns about cloud security are nothing new. Many industries that use HPC heavily have stringent security requirements, and it’s a commonly cited obstacle to cloud-based HPC solutions. 43% of participants in the HPC market survey had concerns about data security and governance, and 42% also listed data privacy. While some perceive security and privacy benefits to on-premises HPC, they don’t account for risk management issues like aging infrastructure that increase the security costs of maintaining compliance, or for expensive regulatory compliance and certifications often required for on-premises solutions. They may not sufficiently account for the complex landscape that security compliance has become and the benefits of a shared responsibility model where AWS helps relieve the customer’s operational burden by operating, managing and controlling the components from the host operating system and virtualization layer down to the physical security of the facilities in which the service operates.

Cloud security is a high priority at AWS and we offer several tools and services to ensure encryption, manage access, and secure regulated workloads. All data is stored in highly secure AWS data centers, and the network architecture is built to meet the requirements of the most risk-sensitive organizations. Additionally, customers maintain ownership and control of all content—they can select which AWS services can process, store and host content, determine where it will be stored, choose its secured state, and manage all access. When a customer uses AWS services, they operate in a shared responsibility environment, where the secure functioning of an application on AWS requires action from both the customer and AWS. All institutions should explain the shared responsibility model to their stakeholders throughout the design, development, testing, and production phases of cloud adoption. Customers are responsible for security IN the cloud. They control and manage the security of their content, applications, systems, and networks. AWS manages security OF the cloud to protect infrastructure and services, maintain operational performance, and meet relevant legal and regulatory requirements. Intel Xeon Scalable processor-based AWS instances deliver hardware-enabled security capabilities directly on the silicon to help protect every layer of the compute stack, including hardware, firmware, operating systems, applications, networks, and the cloud. Intel Threat Detection Technology (Intel TDT) is also available on 2nd Generation Intel Xeon Scalable processors and delivers hardware-enhanced threat detection.

 

Data Transfer

Running HPC applications in the cloud starts with moving the required data into the cloud, but this process can be an obstacle for many organizations. The market study found that 41% of those surveyed were concerned about getting data into—and out of—the cloud. Commonly cited data transfer barriers are time and money, but while it may seem easier in the long run to keep data in on-premises HPC infrastructure, the investment in moving data to the cloud is far outweighed by the benefits of more flexible, agile HPC. Moving data and HPC to the cloud improves efficiency by freeing up valuable financial and staff resources, and reduces business risks by storing data in a more resilient, secure environment. In addition, cloud-based HPC enables customers to utilize AI, machine learning and deep learning to mine all the data available from HPC simulations, narrowing the range of simulations required—meaning cheaper, faster HPC workload execution. Since newer, cloud-native HPC applications were designed to perform better on cloud-based elastic infrastructure, improved performance in the cloud can deliver better ROI.

 

Performance

Organizations using High Performance Computing expect high performance—and many still don’t believe the cloud can compete with on-premises data centers. 35% of participants in the market survey mentioned concerns about network performance and inter-connect latencies, and 29% mentioned broader performance concerns. But the belief that the networking speed between compute nodes in the cloud is not fast enough for high performance is outdated. Recent advancements have helped speed up cloud networking and trim latency to the point where all but the most resource-intensive HPC applications run just as well or better on the cloud than on on-premises infrastructure. AWS performance exceeds the needs of almost every HPC use-case in terms of scalability, elasticity and raw performance, and typically delivers better ROI. Elastic Fabric Adapter (EFA) – a network interface for Amazon EC2 instances offers a unique OS bypass networking mechanism to provides a low latency, low-jitter channel for inter-instance communications. This enables tightly coupled HPC or distributed machine learning applications to scale to thousands of cores, so applications run faster. For a standard CFD simulation, the use of EFA shows a 4x improvement in scaling over using the standard networking for EC2 instances.

In another performance benchmark exercise, Amazon EC2 C5n instances were compared to a mainstream HPC node from a leading on-premises HPC OEM running a standard CFD use case. Engineering simulation software provider ANSYS publishes ANSYS® Fluent® benchmarks of “External Flow Over a Formula-1 Race Car.” This case has around 140-million Hex-core cells and uses the realizable k-epsilon turbulence model as well as the Pressure-based coupled solver and the Least Squares cell-based, pseudo-transient solver. Running the same benchmark using Amazon EC2 C5n instances and Elastic Fabric Adapter is a simple way to benchmark the performance of the solver on AWS and compare it against traditional HPC infrastructure.

The plot below shows the rating of the on-premises OEM’s HPC node and C5n.18xlarge with EFA. ANSYS defines† this rating as “the primary metric used to report performance results of the Fluent Benchmarks. It is defined as the number of benchmarks that can be run on a given machine (in sequence) in a 24-hour period. It is computed by dividing the number of seconds in a day (86,400 seconds) by the number of seconds required to run the benchmark. A higher rating means better performance.”

The plot shows C5n.18xlarge with EFA got a higher rating up to 2400 cores and is essentially on par up to about 3800 cores.

  

Conclusion

HPC is an essential function for many industries, but misconceptions about cloud-based HPC may prevent organizations from realizing the benefits of these powerful systems—like quicker time to market, new business insights, unprecedented agility and scalability, and more. When comparing on-premises infrastructure to cloud-based HPC, it’s important to consider factors beyond a simple cost-per-core-hour analysis and look at the holistic business impact. Factors like personnel productivity, cutting-edge technology, and innovation acceleration are critical in the new digital economy—the difference between leading the industry or playing catch-up.

The ability to create customized compute clusters in the AWS cloud enables cost-effective HPC for most business cases, from small research teams to large enterprise organizations. And AWS offers a suite of integrated products and services that keep data private and secure, make it easy to migrate and transfer data, and deliver consistent high performance. Download the entire Whitepaper to learn more.

Return to Solution Channel Homepage
Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, code-named Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from its predecessors, including the red-hot H100 and A100 GPUs. Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. While Nvidia may not spring to mind when thinking of the quant Read more…

2024 Winter Classic: Meet the HPE Mentors

March 18, 2024

The latest installment of the 2024 Winter Classic Studio Update Show features our interview with the HPE mentor team who introduced our student teams to the joys (and potential sorrows) of the HPL (LINPACK) and accompany Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the field was normalized for boys in 1969 when the Apollo 11 missi Read more…

Apple Buys DarwinAI Deepening its AI Push According to Report

March 14, 2024

Apple has purchased Canadian AI startup DarwinAI according to a Bloomberg report today. Apparently the deal was done early this year but still hasn’t been publicly announced according to the report. Apple is preparing Read more…

Survey of Rapid Training Methods for Neural Networks

March 14, 2024

Artificial neural networks are computing systems with interconnected layers that process and learn from data. During training, neural networks utilize optimization algorithms to iteratively refine their parameters until Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, code-named Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Survey of Rapid Training Methods for Neural Networks

March 14, 2024

Artificial neural networks are computing systems with interconnected layers that process and learn from data. During training, neural networks utilize optimizat Read more…

PASQAL Issues Roadmap to 10,000 Qubits in 2026 and Fault Tolerance in 2028

March 13, 2024

Paris-based PASQAL, a developer of neutral atom-based quantum computers, yesterday issued a roadmap for delivering systems with 10,000 physical qubits in 2026 a Read more…

India Is an AI Powerhouse Waiting to Happen, but Challenges Await

March 12, 2024

The Indian government is pushing full speed ahead to make the country an attractive technology base, especially in the hot fields of AI and semiconductors, but Read more…

Charles Tahan Exits National Quantum Coordination Office

March 12, 2024

(March 1, 2024) My first official day at the White House Office of Science and Technology Policy (OSTP) was June 15, 2020, during the depths of the COVID-19 loc Read more…

AI Bias In the Spotlight On International Women’s Day

March 11, 2024

What impact does AI bias have on women and girls? What can people do to increase female participation in the AI field? These are some of the questions the tech Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Analyst Panel Says Take the Quantum Computing Plunge Now…

November 27, 2023

Should you start exploring quantum computing? Yes, said a panel of analysts convened at Tabor Communications HPC and AI on Wall Street conference earlier this y Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Training of 1-Trillion Parameter Scientific AI Begins

November 13, 2023

A US national lab has started training a massive AI brain that could ultimately become the must-have computing resource for scientific researchers. Argonne N Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire