This article provides a comprehensive overview of the Lustre parallel file system, covering its architecture, key features, component roles, scalability, performance characteristics, and step‑by‑step configuration procedures for high‑performance computing environments.
This article explains the distinction between distributed systems and clusters, outlines cluster’s key features such as scalability and high availability, describes essential capabilities like load balancing and error recovery, details core technologies, classifies common Linux cluster types, and provides examples to illustrate their operation.
The article provides a comprehensive overview of the Lustre file system, detailing its cluster storage architecture, core components, scalability, performance optimizations, security mechanisms, high‑availability features, and usage in high‑performance computing environments.
This article examines why the traditional Linpack‑based TOP500 list is being complemented by newer benchmarks such as HPCG, Graph500, Green Graph 500 and IO‑500, explains their methodologies, presents the 2017 ranking results for major Chinese supercomputers, and reviews a wide range of application and micro‑benchmarks used to evaluate HPC system performance.
This article explains why enterprises are finally embracing hybrid cloud for high‑performance computing, outlines the security and technical concerns that previously held them back, and provides concrete strategies—such as automation, cloud bursting, workflow preservation, and custom images—to achieve a smooth migration.
The article explains how heterogeneous parallel computing—leveraging CPUs, GPUs, FPGAs and other specialized units—addresses the performance limits of traditional serial programming by distributing tasks across diverse hardware, detailing its concepts, architectures, development models, and relevance to AI and cloud workloads.
This article explains the concepts of MPI, OpenMPI, and OpenMP, compares three major server architectures—SMP, NUMA, and MPP—and discusses their performance characteristics, scalability limits, and typical application scenarios in high‑performance computing.
The article provides a comprehensive overview of high‑performance computing applications, classifying cluster types, describing computing‑intensive, data‑intensive and network‑intensive workloads, and detailing specific domains such as CAE, CFD, life‑science, animation rendering and meteorology along with resource‑allocation guidance.
This article provides a comprehensive overview of high‑performance computing, covering its rapid market growth, definition, classification into high‑throughput and distributed computing, key hardware components such as CPUs, GPUs, memory types, networking technologies like InfiniBand, performance metrics, benchmarking tools, and parallel file systems.
This article reviews recent HPC market size and growth trends, highlights AI‑driven investment, and evaluates Huawei’s hardware, cloud, and industry‑specific solutions—including supercomputing, automotive, oil & gas, and entertainment—while offering personal insights on storage challenges and future directions.
The article examines the shifting landscape of high‑performance computing storage, detailing how Intel’s abandonment of Lustre, the rise of BeeGFS, and evolving demands from large‑scale analytics and AI are driving organizations to adopt more flexible, user‑space parallel file systems that promise better performance, scalability, and easier management.
The article provides a detailed market analysis of high‑performance computing from 2016 to 2021, covering market size, budget allocation, regional shares, key applications, the rise of AI and cloud integration, and emerging hardware and networking technologies driving future growth.
The article surveys how NVIDIA's GPU ecosystem and the Intersect360 study reveal that among the 50 most common high‑performance computing applications, 34 already support GPU acceleration, highlighting the growing importance of GPU computing across scientific, engineering, and business fields.
This article provides a comprehensive overview of IBM GPFS, covering its architecture, management components, networking models, cluster and storage design, as well as practical guidance on building‑block configurations for performance and capacity scaling in high‑performance computing environments.
The article examines the 2017 TOP500, Green500, HPCG, Graph500 and Green Graph 500 rankings, explains why Linpack is becoming insufficient, compares benchmark methodologies, and introduces major application and micro‑benchmarks that illustrate the evolving performance metrics of modern supercomputers.
This article compares NVIDIA's Tesla and GeForce GPU families, detailing their target markets, design differences, core architectures, double‑precision performance, ECC support, memory bandwidth, interface options, software and OS compatibility, power efficiency, and management features to help readers choose the right GPU for HPC or gaming workloads.
BeeGFS is a GPL‑licensed parallel file system for Linux that offers scalable storage through a modular architecture of management, metadata, and object storage servers, supports a wide range of hardware and OS platforms, and provides detailed installation, configuration, and performance‑tuning guidance including the BeeOND burst‑buffer extension.
The 2017 TOP500 list shows China's Sunway TaihuLight leading with 125,435.9 TFlop/s, highlights the role of Linpack‑based rankings, details vendor participation, and contrasts performance‑focused TOP500 with energy‑focused Green500, underscoring China's growing high‑performance computing capabilities and the importance of power efficiency.
The article analyzes how expanding demand and cross‑industry convergence are driving High‑Performance Computing (HPC) to become the next technology hotspot, covering its market segmentation, cloud integration, storage challenges, key vendors, and future trends such as HPDA, Burst Buffer, and open‑source file systems.
In an interview at the 2015 HPC User Forum, experts discuss how cloud computing, big data analytics (HPDA), deep learning, and software‑defined architectures are reshaping high‑performance computing and outline emerging trends, challenges, and strategic advice for Chinese HPC vendors.
This article provides a comprehensive overview of the Lustre parallel distributed file system, detailing its architecture, stripe configuration, I/O performance traits, challenges with small files, and practical optimization techniques for high‑performance computing environments.
Intel announced it will cease providing its proprietary Lustre file‑system version, contribute all features to the open‑source community, designate the community build as the long‑term‑stable release, and release related Hadoop and HPC adapters, signaling a strategic shift in its high‑performance‑computing operations.
The article explains what Burst Buffer technology is, how NERSC integrates Cray DataWarp‑based flash/SSD buffers into its Cori supercomputer to boost I/O bandwidth and IOPS for scientific workloads, and describes the architecture, software stack, performance benefits, roadmap, and competing solutions from other vendors.
The article reviews the evolution of Direct Attached Storage (DAS) from its early role as basic direct‑connected storage to its renewed relevance in big‑data, cloud, and HPC environments, detailing market growth, vendor offerings, architectural designs, and practical deployment scenarios.
This article explains the core differences between distributed systems and clusters, detailing their architectures, efficiency goals, typical use cases such as Hadoop MapReduce and load‑balancing clusters, and outlines key concepts like scalability, high availability, load balancing, and error recovery.
This article explains what a Linux cluster is, contrasts it with distributed systems, outlines its two main characteristics—scalability and high availability—along with essential capabilities like load balancing and error recovery, and details common cluster types such as high‑availability, load‑balancing, and high‑performance computing clusters.
