UW-Madison Logo

The ADvanced Systems Laboratory (ADSL)
Publication abstract

Datamation 2001: A Sorting Odyssey

Florentina I. Popovici, John Bent, Brian C. Forney, Andrea C. Arpaci-Dusseau, and Remzi H. Arpaci-Dusseau
UW Technical Report CS-TR-2002-1444
August 2002

Abstract:

We present our experience of turning a Linux cluster into a high-performance parallel sorting system. Our implementation, WiND-Sort, broke the Datamation record by roughly a factor of two, sorting 1 million 100-byte records in 0.48 seconds. We have identified three keys to our success: developing a fast remote execution service, configuring the cluster properly, and avoiding the potential ill-effects of occasionally faulty hardware.

Full Paper: Postscript   PDF