The development of microscopic techniques makes the data structure ever more complicated. In many cases, the raw data is large, time consuming to process and not needed at the end of the experiment. This module is designed to simplify the workflow to improve usability and to parallelize the acquisition and processing steps to save time.
During image acquisition using an acquisition PC, Direct Processing enables the user to select processing functions which are executed on a processing PC. Several different functions are available and you can define a sequence of functions (in a so-called pipeline), which are then executed sequentially one after another. Direct Processing starts to process the smallest processable entity as soon as its acquisition has been completed. In the case of deconvolution, this is typically a z-stack for one channel.
Direct Processing allows an acquisition computer to communicate with a second PC (processing computer) connected via a network connection. The acquisition computer instructs the processing computer to process images as they are being acquired. Before you can use the module, you typically need to connect the two computers. As there are numerous ways, how computers can be set up to become networked, we can only give some general advice here. Contact your local IT administration for help on how to configure the computers in accordance to the local infrastructure.
One way is to connect the computer controlling the microscope system to a second computer via a direct ethernet connection. If Network Discovery is switched on, Windows 10 will directly support such a point to point connection. Create a shared folder which can be accessed from both computers. Since no other network traffic will use this connection, the whole bandwidth will be available for saving data directly from the acquisition computer to a shared folder on the processing computer. This is the most efficient way as the acquired data do not have to be copied off the acquisition computer after the acquisition has finished.
Alternatively, it is also possible, to use a processing computer which is already integrated into an existing network. Depending on the network type and the kind of experiments being done, the bandwidth might not be sufficient to directly stream data to the processing computer. In such cases, to not limit the throughput of the acquisition, it is advisable to let the acquisition computer acquire data to a local drive and instruct the processing computer where to look for the acquired data.
If no processing PC is available, it is also possible to configure the same acquisition PC for parallel processing steps, i.e. acquire and process images on the same workstation.
For a communication via network, it is possible to link both computers with a discovery proxy. The discovery proxy is a service where you can register all available processing computers and your acquisition computer can then ask this service for the list of these available PCs. You can also link the PCs without such a discovery service, for which you then need the IP address and/or network name of the processing computer. Note that such a use of a discovery proxy is primarily designed for configurations in which you have more than one processing PCs.