The Quadrics network (QsNet): high-performance clustering technology

• Published in: HOT 9 Interconnects. Symposium on High Performance Interconnects pp. 125 - 130
• Main Authors: Petrini, F., Wu-chun Feng, Hoisie, A., Coll, S., Frachtenberg, E.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2001
• Subjects: Fault detection; Fault tolerance; Hardware; Multiprocessor interconnection networks; Network interfaces; Network operating systems; Operating systems; Protocols; Radio access networks; Technological innovation
• ISBN: 0769513573; 9780769513577
• DOI: 10.1109/HIS.2001.946704

The Quadrics interconnection network (QsNet) contributes two novel innovations to the field of high-performance interconnects: (I) integration of the virtual-address spaces of individual nodes into a single, global, virtual-address space and (2) network fault tolerance via link-level and end-to-end protocols that can detect faults and automatically re-transmit packets. QsNet achieves these feats by extending the native operating system in the nodes with a network operating system and specialized hardware support in the network interface. As these and other important features of QsNet can be found in the InfiniBand specification, QsNet can be viewed as a precursor to InfiniBand. In this paper, we present an initial performance evaluation of QsNet. We first describe the main hardware and software features of QsNet, followed by the results of benchmarks that we ran on our experimental, Intel-based, Linux cluster built around QsNet. Our initial analysis indicates that QsNet performs remarkably well, e.g., user-level latency under 2 /spl mu/s and bandwidth over 300 MB/s.