Nsort is a sort/merge program that can quickly sort large amounts of data, using large numbers of processors and disks in parallel. Unique in its CPU efficiency, Nsort is the only commercial sort program to demonstrate:

• 1 Terabyte sorts (33 minutes)
• 1 Gigabyte/sec file read and write rates

Nsort has a long history of sorting massive, production data sets, such as:

• Web logs for high-traffic web sites
• Phone logs
• Government agency data

Nsort is integrated into Informatica Corporation's analytic software products, used by over 4,440 customers including 84 percent of the Fortune 100.

Nsort for SSIS is a high-speed sort compoment for Microsoft's SQL Server Integrations Services based on the Nsort subroutine library, and can sort arbitrarily large datasets.

Nsort is the sort engine behind Fujitsu Software's NeoSort program which allows mainframe sorts to be executed on Windows systems while multiple processors and disks.

Nsort can be used in conjunction with SUMP Pump, a open source project that enables the rapid development of multithreaded code for processing streams of data both before and after sorting. Together, Nsort and SUMP Pump allow Hadoop streaming scripts to be run on a single node using all available CMP/SMP processors and single input and output files.

The Nsort library can be used by SPSS server 15.0 and beyond, providing SPSS users with the performance benefits of a multiprocessor sort.

You can try out Nsort immediately with our free online trials. Nsort is available on:

• Red Hat Linux (x86 or x64, including a free text-only license option)
• Microsoft Windows (x86 or x64)
• Sun Solaris (SPARC)
• Hewlett-Packard HP-UX (PA-RISC or Itanium)
• AIX (available on request)