Projects: ASCI

References

General

• Infiniband Trade Association
• Mellanox Technologies (Infiniband company)
• Computational Plant
• Myricom
• Myrinet -- A Gigabit-per-Second Local Area Network
• ASCI Purple
  The Statement of Work document on this site has a great overview of the ASCI project, all of the current clusters, types of applications run on the clusters and how NFSv4 will be expected to be used (at the minimum) on the Purple cluster.
• Earth Simulator Project
  This is currently the most powerful computer in operation.

Clusters

• Workload Characterization of Input/Output Intensive Parallel Applications (1997)
• Performance Analysis Of Tape Libraries For Supercomputing Environments
• A Comparison of Logical and Physical Parallel I/O Patterns (1998)
• Parallel I/O
• I/O, Performance Analysis, and Performance Data Immersion (1996)
• Processing Large-Scale Multidimensional Data in Parallel and Distributed Environments
• Glossary on Parallel Input/Output
• On Implementing MPI-IO Portably and with High Performance (1999)
• Benchmark Design for Characterization of Balanced High-Performance Architectures (2001)
• Integrating Parallel File I/O and Database Support for High-Performance Scientific Data Management (2000)
• Structure and Performance of the Direct Access File System
• I/O in Parallel Applications: The Weakest Link
• Optimizing Noncontiguous Accesses in MPI-IO
• Parallel I/O Archive
• SUN MPI I/O: Efficient I/O for Parallel Applications
• ROMIO: A High-Performance, Portable MPI-IO Implementation
• MPI Forum (MPI-2.0 Documentation with MPI-IO Spec)
• MPI/ROMIO Documents
• Performance Evaluation of Parallel I/O in Cluster Environments
• The Parallel Effective I/O Bandwidth Benchmark: b_eff_io
• The parallel communication and I/O bandwidth benchmarks: b eff and b eff io (2001)
• Dynamic I/O Characterization of I/O Intensive Scientific Applications
• Dynamic File-Access Characteristics of a Production Parallel Scientific Workload
• I/O Requirements of Scientific Applications: An Evolutionary View (1996)
• Tuning the Performance of I/O-Intensive Parallel Applications
• Exploiting Inter-File Access Patterns Using Multi-Collective I/O

Distributed Systems

• A Fast Mutual Exclusion Algorithm (1986)
• Local-Area Distributed Systems.
• Parallel Communication in a Large Distributed Environment
• Exploiting Gray-Box Knowledge of Buffer-Cache Contents
• Topologically-Aware Overlay Construction and Server Selection
• The Dangers of Replication and a Solution (1996)
• A Two-Level Checkpoint Algorithm in a Highly-Available Parallel Single Level Store System
• IBM's Open Source Distributed Lock Manager Project
• Recovery and Coherency-Control Protocols for Fast Intersystem Page Transfer and Fine-Granularity Locking in a Shared Disks Transaction Environment.

Memory

• Memory Coherence in Shared Virtual Memory Systems
• Techniques for Reducing Consistency-Related Communication in Distributed Shared
• Cooperative Caching Middleware for Cluster-Based Servers (2001)

File Systems

• The NFS Version 4 Protocol (2000)
• Sistina GFS
• Lustre
• GPFS Product Page
• GPFS Research Page
• MPI-IO/GPFS, an Optimized Implementation of MPI-IO on top of GPFS
• GPFS: I/O Scalability and Availability for AIX and Linux Clusters
• Linux Clustering With CSM (Client Systems Management) and GPFS
• Towards a High-Performance and Robust Implementation of MPI-IO on top of GPFS
• Distributed file systems: concepts and examples
• NFS Version 3 Design and Implementation
• Synchronization and Caching Issues in the Andrew File System
• Scale and Performance in a Distributed File System
• DEcorum File System Architectural Overview
• Sailesh Chutani , B.A. Bottos, DEcorum Token Manager Interface Specification, Transarc Corporation
• DFS TOKEN MANAGER REDESIGN
• The Episode File System (1992)
• Coda: A Highly Available File System for a Distributed Workstation Environment
• Bruce J. Walker and Stephen H. Kiser, "The LOCUS Distributed Filesystem," in The LOCUS Distributed System Architecture, ed. Gerald Popek and Bruce J. Walker, The MIT Press Series in Computer Systems, Herb Schwetman, gen. ed., The MIT Press (1985).
• Caching in the Sprite Network File System (1988)
• A Comparison of Three Distributed File System Architectures: Vnode, Sprite, and Plan 9
• Scalability in the XFS File System
• Parallel Virtual File System
• Ocean Store Website
• * OceanStore: An Architecture for Global-Scale Persistent Storage
• LDFS: A Fault-Tolerant Local Disk-Based File System for Mobile Agents
• Removing Bottlenecks in Distributed Filesystems: Coda & InterMezzo as examples
• Galley: A New Parallel File System For Scientific Workloads (96)
• Performance of the Galley Parallel File System (1996)
• File-Access Characteristics of Parallel Scientific Workloads (1996)
• RAMA: An easy-to-use, high-performance parallel file system
• Sun's Parallel File System (PFS)
• Sun HPC ClusterTools 4: Administrator's Guide

Load Balancing

• Dynamic Function Placement for Data-intensive Cluster Computing (2000)
• Interposed Request Routing for Scalable Network Storage

Storage

• Demystifying Storage Networking
• Network Attached Storage Architecture
• Storage Area Networks and the High Performance Storage System (2002)
• About SANs.com
  This is a useful site with lots of information about Storage Area Networks including Whitepapers, Articles, News, etc

Benchmarks

• MCNP - A General Monte Carlo N-Particle Transport Code
• The ASCI Purple Benchmark Codes
  sPPM, UMT1.2, SMG2000 and SAGE have been recommended to us by LLNL researchers. A MPI-IO benchmark and POSIX file I/O benchmark are also listed on this page.
• FLASH (You need to search for it)
• Radiation Safety Information Computational Center
  This has a list of various code packages.