Inside HPC & AI News | High-Performance Computing & Artificial Intelligence
At the Convergence of HPC, AI at Scale, Quantum
Subscribe
  • News
    • AI News
    • Business of HPC
    • New Installations
  • HPC-AI Hardware
    • Compute
    • CPUs, GPUs, FPGAs
    • Exascale
    • Future Technology
    • Green HPC
    • HPC/AI Chips and Systems
    • Network
    • Quantum Computing
    • Storage
  • HPC-AI Software
    • AI & Machine Learning
    • Cloud HPC
    • High Performance Analytics
    • Lustre
    • Parallel Programming
    • Systems Management
    • Tools
  • Quantum
  • Resources
    • Thought Leader Articles
    • Education / Training
    • Events Calendar
    • HPC Career Notes
    • Industry Perspectives
    • Industry Segments
      • Enterprise HPC
      • Financial Services
      • Government
      • Manufacturing
      • National Lab News
      • Research / Education
    • Jobs Board
    • Vanguards HPC-AI
    • Special Reports
    • The Exascale Report Archives
    • White Papers
  • Podcasts & Videos
    • @HPCpodcast
    • Other Podcasts
    • Videos
  • Power & Cooling
    • Advanced Tech & Efficiency
    • Air & Liquid Cooling
    • Data Center
    • Green Data Center
    • Infrastructure Design/Management
    • Interconnects & Networking
    • Nuclear, Solar, Wind, LNG, Geothermal, Fusion
    • Sustainability
    • System & Facility Monitoring
  • AI News
  • Search

Video: Scalability Testing of DNE2 in Lustre 2.7

May 4, 2015 by Doug Black
Print Friendly, PDF & Email
  • share 
  • share 
  • share  
  • share  
  • email 

In this video from LUG 2015 in Denver, Steve Simms from Indiana University presents: Scalability Testing of DNE2 in Lustre 2.7.

The Lustre Distributed Namespace (DNE) feature allows for the distribution of file and directory metadata across multiple Lustre Metadata Targets (MDT). With this, you can now effectively scale your metadata load across a near unlimited number of targets. This both helps in reducing single metadata server hardware requirements as well as greatly expanding the maximum number of files your Lustre file system may contain. DNE Phase 2 enables deployment of striped directories on multiple MDS nodes and improved versatility.

Increasing Lustre’s metadata performance is something that Indiana University HPC users greatly desire. Because of the many user comments and requests, the High Performance File Systems group at IU has been looking into metadata performance using solid state storage devices. The most recent tests that we have performed involved the use of multiple metadata servers and the striped directory functionality provided by DNE2. This presentation will feature the data we have gathered on the relationship between metadata performance and MDT count using DNE2.”

Download the Slides * See more talks in the LUG 2015 Video Gallery. * Sign up for our insideHPC Newsletter.

  • share 
  • share 
  • share  
  • share  
  • email 
Filed Under: Events, HPC-AI Software, Lustre, Resources, Videos Tagged With: DNE, LUG 2015, Metadata
«
»
»
«

Sponsored Guest Articles

Unleashing Power: NVIDIA L40S Data Center GPU by PNY

The NVIDIA L40S Data Center GPU, provided by PNY, represents a significant leap forward in the realm of high-performance computing. This GPU is engineered to meet the demanding needs of modern data centers ….

White Papers

Lamini Chooses SuperMicro GPU Servers for LLM Tuning Offering

Lamini is developing an infrastructure for customers to run Large Language Models (LLMs) on innovative and fast servers. End-user customers can use Lamini’s LLMs or build their own using Python, an open-source programming language. Lamini has developed a software environment for customers that allows them to focus on their business needs and develop innovative AI […]

Download
More White Papers

Join Us On Social Media

Featured From
  • DDN Report: 65% of Organizations Struggling to Achieve AI Success

    The findings reveal that while organizations are seeking to adopt AI, most lack the foundation to sustain it. Rising complexity associated with infrastructure, underutilized cloud ….

More News from insideAI News

  • HPC News Bytes 20260112: Big Chip News at CES, TSMC’s Arizona Expansion, Sandia’s Neuromorphic News
  • CIQ Delivers Turnkey Sovereign AI with Service Endpoints
  • DDN Powers Integrated Compute, Data, and Offload at Scale for NVIDIA Rubin Platform
  • VAST Data Redesigns Inference Architecture for Agentic AI with NVIDIA
  • Marvell to Acquire XConn for AI Data Center Connectivity
  • Supermicro Expands Edge Portfolio with NVIDIA, Intel and AMD-Powered Platforms
  • NVIDIA Releases Details on Next-Gen Vera Rubin AI Platform — 5X the Performance of Blackwell
  • About insideHPC
  • Contact
  • Advertise with insideHPC
  • Visit Our Other Site – insideBIGDATA
  • Terms of Service & Copyright
  • Privacy Policy
Inside HPC & AI News | High-Performance Computing & Artificial Intelligence
Copyright © 2026