Faster MathLink
Mathematica 5 uses optimized TCP/IP protocol for MathLink communication between
front end and kernel, and between multiple kernels such as for gridMathematica deployments.
These communications now occur at the speed of the underlying network.
The optimized TCP/IP protocol improves bandwidth by a factor of 10 and
latency by a factor of 200 on a standard 100Base-T network. With
gigabit networks and crossbars found in leading-edge computing
centers, the speed-ups are even greater. Other changes--for example, a
new shared memory protocol for Windows platforms--give a
10-fold improvement for communication between parts
of Mathematica on the same machine.
J/Link, .NET/Link, and most import
and export
formats rely on MathLink, thereby benefitting from performance
increases in MathLink.
|