
Software
LabVIEW driver and examples
(c) Spectrum Instrumentation GmbH
61
LabVIEW driver and examples
A full set of drivers and examples is available for LabVIEW for Windows. Lab
-
VIEW for Linux is currently not supported. The LabVIEW drivers have their own
manual. The LabVIEW drivers, examples and the manual are found on the CD
that has been included in the delivery. The latest version is also available on our
webpage www.spectrum-instrumentation.com
Please follow the description in the LabVIEW manual for installation and useage
of the LabVIEW drivers for this card.
MATLAB driver and examples
A full set of drivers and examples is available for Mathworks MATLAB for Windows (32 bit
and 64 bit versions) and also for MATLAB for Linux (64 bit version). There is no additional
toolbox needed to run the MATLAB examples and drivers.
The MATLAB drivers have their own manual. The MATLAB drivers, examples and the manual
are found on the CD that has been included in the delivery. The latest version is also avail
-
able on our webpage www.spectrum-instrumentation.com
Please follow the description in the MATLAB manual for installation and useage of the
MATLAB drivers for this card.
SCAPP – CUDA GPU based data processing
Spectrum’s CUDA Access for Parallel Processing
Modern GPUs (Graphic Processing Units) are de
-
signed to handle a large number of parallel opera
-
tions. While a CPU offers only a few cores for
parallel calculations, a GPU can offer thousands of
cores. This computing capabilities can be used for
calculations using the Nvidia CUDA interface.
Since bus bandwidth and CPU power are often a
bottleneck in calculations, CUDA Remote Direct
Memory Access (RDMA) can be used to directly transfer data from/to a Spectrum Digitizer/Generator to/from a GPU card for processing,
thus avoiding the transfer of raw data to the host memory and benefiting from the computational power of the GPU.
For applications requiring high performance signal and data processing Spectrum of
-
fers SCAPP
(Spectrum’s CUDA Access for Parallel Processing). The SCAPP SDK allows a direct link
between Spectrum digitizers or generators and CUDA based GPU cards. Once in the
GPU users can harness the processing power of the GPU’s multiple (up to 5000) pro
-
cessing cores and large (up to 24 GB) memories. SCAPP uses an RDMA (Linux only)
process to send data at the digitizers full PCIe transfer speed to the GPU card. The SDK
includes a set of examples for interaction between the digitizer or generator and the
GPU card and another set of CUDA parallel processing examples with easy building
blocks for basic functions like filtering, averaging, data de-multiplexing, data conver
-
sion or FFT. All the software is based on C/C++ and can easily be implemented, ex
-
panded and modified with normal programming skills.
Please follow the description in the SCAPP manual for installation and usage of the SCAPP drivers for this card.
Image 42: LabVIEW driver oscilloscope example
Image 43: Spectrum MATLAB driver structure
Image 44: GPU usage with SCAPP SDK: data transfer options
Summary of Contents for M2p.59 Series
Page 190: ......