| Publisher | Sun Microsystems | ||
|---|---|---|---|
| Format | 927.8KB PDF | Date added | 01 Feb 2008 |
| Topics | Scalability, File and Network Servers | ||
| Downloads | 119 | ||
This paper describes low-level infrastructure in the Lustre file system that addresses scalability in very large clusters. The features described deal with I/O and networking, lock management, recovery after failure, and other scalability-related issues. The Lustre file system first went into production in Spring 2003 on the Multiprogrammatic Capability Resource (MCR) cluster at Lawrence Livermore National Laboratory (LLNL). The MCR cluster was one of the largest clusters at that time with 1100 Linux compute nodes as Lustre clients. Since then, the Lustre file system has been deployed on larger systems, notably the Sandia Red Storm deployment, with approximately 25,000 liblustre clients, and the Oak Ridge National Laboratory (ORNL) Jaguar system, which is of similar scale and runs Lustre technology both with Client-Node Linux (CNL) and with Catamount.
Related white papers
Management Strategy for Network-Critical Physical Infrastructure
When choosing a management solution for the physical infrastructure of IT networks, management of individual devices is necessary in order to have visibility to the many data points required for...
Protecting Data With a Scalable, Flexible, and High-Performance Solution From Symantec
Following the merger of Halifax and Bank of Scotland in 2001, HBOS, the merged entity of Halifax and Bank of Scotland, has 22 million customers and assets of more than...
HP Technology Helps Lutz Grow and Adapt to Change Over Time
Transforming a small family business into Europe's largest home furnishings retailer requires keen business acumen and the ability to manage change over time. Founded in 1945, Lutz began as a...
Multinational Healthcare Provider Simplifies Platforms and Networking With HP ProLiant Servers and HP NC7170 Network Adapters
Founded in 1971 in Columbus, Ohio, Cardinal Health has grown into a multinational healthcare provider for customers ranging from healthcare manufacturers to patient care providers. Cardinal Health has evolved beyond...
Take a Multimedia Tour of Intel's I/O Acceleration Technology
You can't address the root causes of slow application response times merely by increasing your bandwidth. The challenge is to increase CPU efficiency and speed up TCP/IP processing, and Intel's...
VL2: A Scalable and Flexible Data Center Network
To be agile and cost effective, data centers should allow dynamic resource allocation across large server pools. In particular, the data center network should enable any server to be assigned...
A Path Availability Model for Wireless Ad-Hoc Networks
Ad-hoc networks are expected to play an important role in future commercial and military communications systems. As such, scalable routing strategies capable of supporting greater user mobility and a wide...




