Today’s HPC systems are being used to solve complex problems in industry and research or academia that were previously unsolvable in many areas such as nanotechnology, climate change, neuroscience, bioinformatics, computational biology and astrophysics. As a result, more researchers and scientists than ever before are publishing papers that detail their advanced research. This research is often based exclusively on simulations obtained on HPC Systems.
Hewlett Packard Enterprise (HPE) is uniquely positioned to fill this new discipline because our broad range of servers are purpose-built for scale and efficiency and are highly adaptable to changing requirements and technology. That means that an HPE solution can be a perfect fit for almost any size organization and any challenge. HPE customers include a wide range of organizations including smaller research institutes, university computing centers, regional HPC centers, and private or governmental research entities.
Products Portfolio of HPC and Big Data
HPE’s broad range of products provides the science and research community with more performance in less space with less power consumption. More performance means that researchers and scientists can improve parallelism to run bigger applications more thoroughly-now and in the future.
The HPE Apollo Systems Family Apollo 2000 System: the bridge to enterprise scale-out architecture Deploy more compute power to reap the benefits of cloud business, deploy Web-based applications, and increase HPC power to speed research and bring new products and services to market faster—within space and resource boundaries. The Apollo 2000 System provides a bridge to scale-out architecture for traditional data centers so you can achieve the space and cost savings of density-optimized infrastructure in a non-disruptive manner.
Apollo 4000 System Portfolio: Server solutions purpose-built for Big Data This new family is purpose-built to service the technologies that are driving the Big Data IT revolution—Hadoop and other Big Data analytics solutions and object storage systems. These purpose-built systems will give you a competitive edge for accelerated market share and business growth by overcoming your data center challenges of space, energy, and time.
A recently announced member of the Apollo 4000 is the Apollo 4520 System; Lower total cost of ownership with new HPC storage server solution. Designed to support Lustre implementations – including the HPE supported Lustre solution, based on the Intel® Enterprise Edition of Lustre.
Apollo 6000 System: rack-scale solutions with better density, performance, power efficiency, and cost of ownership To address the growing demand for HPC and the relentless pursuit of efficiency, Hewlett Packard Enterprise has taken the lead on a new approach: thinking beyond just the server and designing a rack‑level solution that gives you the right compute at the right economics so you can get the most out of your infrastructure—and your budget. The HPE Apollo 6000 System delivers industry‑leading performance in less space with the flexibility to tailor the system to precisely meet workload requirements.
Apollo 8000 System: advancing the science of supercomputing The HPE Apollo 8000 System is an HPC solution that uses innovative warm-liquid cooling technology to fuel the future of science and technology—with a “green” answer to some of the industry’s toughest challenges. The HPE Apollo 8000 System reaches new heights of performance density while actually helping reduce your carbon footprint and by providing blade-like serviceability.
The Apollo family is complemented by HPE’s widely known Proliant DL-series of rack-based servers and the BL-series of blade-severs.
HPE Sites with Large-scale parallel file-systems in production
- Tokio Institute of Technology (Japan)
- Purdue University (USA)
- Cyfronet Poland
- University of Southern California (USA)
- Indian Institute of Technology
- Center for Biological Sequence Analysis – DTU (Denmark)
- National Supercomputing Center Sweden
- CSC Finland
Customer involvement and collaboration
HP-CAST, the High Performance – Consortium for Advanced Scientific and Technical Computing is HPE’s users group that works to increase the capabilities of HPE solutions for large-scale, scientific and technical computing. HP-CAST provides a powerful forum for users to relay their experiences with early HPE systems, influencing the development of future products.
HP-CAST provides two world-wide conferences per year, associated with ISC in Europa and SC in the United States and numerous regional conferences.
HP-CAST meetings typically include corporate briefings and presentations by HPE executives and technical staff (under NDA), and discussions of customer issues related to high-performance technical computing. The HP-CAST website is an excellent resource providing information about current and past HP-CAST user group conferences and meetings. For more information see http://www.hp-cast.org or http://www.hp.com/go/hpcast.