common

output

standard output of your code is displayed in left pane.

your request is queued

you may experience latency if service is busy.

execution

limited time execution: your program is killed if amount of real time exceeds the duration you selected.

for free computing the maximum duration is 60 secondes.

the system allocates to your job a private temporary directory with full read/write access.

scilab

modules

external modules are not available yet except sciGPGPU

sciGPGPU module

module sciGPGPU is automatically loaded if you request at least one GPU (intensive computing ony).

if needed kernel files must be present on same directory that scilab source file.

.... = gpuBuild("mykernelfile");

graphic mode

you get back the last figure you drew ex: plot2d()

export function of graphic window 0 is appended to your code

pycuda

modules

list

code

no preloading is done, you must import what you need, example:

import pycuda.autoinit

import pycuda.driver as drv

import numpy

top