Criteria for choosing hardware for Dispatcher Server components

Prior to determining the type of machines to use for translations, determine the Dispatcher Server system usage to set up the correct environment to support the expected throughput. Number of users The number of users submitting translation tasks must be accounted for when trying to get the maximum amount of throughput. If a site has 100 users who can potentially submit 100 simultaneous translation tasks, and the expected throughput is 1 hour, there must be sufficient computational capacity to handle the load. This can be accomplished by providing enough CPU and memory to handle the 100 tasks. Increasing load capacity can also be handled by adding additional modules to handle the additional throughput. As throughput requirements increase or decrease, modules can be started or stopped dynamically to handle the load requirements. Size and complexity of the translation models Knowing the average size of the models that use the system gives an idea as to the minimal requirements required for supporting the normal day-to-day operational needs. Consider the size of the models to be submitted for translation estimates when planning for the number of translations that can be executing simultaneously. Note If a machine has 10 GB memory available, and the average size of the model requires 4 GB of memory to translate, the maximum number of translators that should be configured to run at a given time should be two. Model complexity influences translation times. Models with many curves are more geometrically complex to translate than models with many straight edges. Computational ability of the translation machine The components required to get the data from Teamcenter to the translators are mainly pipeline mechanisms for moving data and communication and require proper network bandwidth for the expected data load, which is unique to each site. The machine performing the actual translation task must have enough CPU, memory, and disk space capacity to load and translate the models. Disk space requirements Understanding the disk space requirements needed ensures that the translations do not run out of disk space during a translation. To obtain the required disk space requirements, multiply the average sized model with the number of possible simultaneous translations. Average model size * number of translations = Average disk space needed at any given time. You also need to account for the generated data, logs, and CAD system files while calculating the disk space needed.