淘宝官方店推荐课程在线工具联系方式关于我们

	微波射频仿真设计 Ansoft Designer 中文培训教程 \| HFSS视频培训教程套装
	Agilent ADS 视频培训教程 \| CST微波工作室视频教程 \| AWR Microwave Office

首页 >> Ansoft Designer >> Ansoft Designer在线帮助文档

Ansoft Designer / Ansys Designer 在线帮助文档：

Remote Analysis >
Remote Configuration Considerations >
Configuring the Local Machine to Solve Remotely

Configuring the Local Machine to Solve Remotely

Configure the local machine to solve remotely by selecting the Analysis Options tab of the Tools > Options > General Options dialog.

The following controls are available.

Analysis Options For Design Type

• Design Type selects the design type for remote analysis. Available types are Nexxim Circuit, Nexxim Netlist, Planar EM, and System.

• To override/specify the machine that you use for remote analysis each time you run a simulation, you can instead check Prompt for analysis machine when launching analysis. Checking this selection displays a Run Simulation On dialog whenever you begin to run a simulation. You can then select either the local machine, or specify a machine name, at the start of each simulation. You can then Save Settings as Default.

Analysis Machine Options

Select whether the default analysis machine should be the Local machine, a Remote machine, or whether analysis should be Distributed across multiple machines.Regardless of the machine(s) on which the analysis is actually run, the number of processors and Desired RAM Limit settings and the default process priority settings are read from the machine from which you launch the analysis. For more information, see Specifying Solution Settings.

• Local — Specifies that the default analysis machine should be the Local machine.

• Remote — If you selected Remote, enter the default analysis machine information either as an IP address, a DNS name (machine_name), or a UNC name (\\machine name).

• Distributed — If you selected Distributed, you can add machines to a list, or edit an existing machine list. Select the Edit button to display the Distributed Analysis Machines dialogue. Here you specify an IP address, a DNS name, or a UNC name for each machine to add to the list. Control buttons let you Add Machine to List to or Remove machines from the list.

Selecting Distributed always checks out the ansoft_distrib license, regardless of whether there is anything distributable or not.If Distributed is selected for a solve of sweeps or distributable Optimetrics, Designer selects machines in a top down order, depending on their load. Designer selects the highest machine on the list that has the least load. Machine load is determined by the number of solve processes running on a machine that were launched within a single session of Designer. For example, if the list contains 4 machines and the topmost machine is already running a simulation, when 2 machines must be selected, machines 2 and 3 in the list are chosen and their loads are updated by 1.

If Distributed is selected when you run a nominal solve, Designer always selects the topmost machine listed.The displayed list always shows the order in which you entered them irrespective of the load on the machines. To control the list order, select one or more machines, and use the Move up or Move down buttons. Click OK to accept the changes and close the Distributed Analysis Machines dialog. The machines are listed in the Analysis Machines Options field in alphabetical order, rather than the distributed list order.

Distributed Machine Configurations

Opens an editable dialog that allows you to Add, Edit, Delete, or Clone the distributed machine configurations listed in the Configuration Name window.

Queue All Simulations

Select Queue all simulations to allow multiple simulations to be added to a queue and run one after another. The local or remote machine specified as the Default Location is used to queue and run the simulations. If you change the local or remote machine specified as the Default Location when the queue is not empty, the next simulation after the one currently running will be launched with the new settings. By default, queuing is disabled.

When all information on the Analysis Options tab is as desired, select another tab or click OK to put your options into effect. For more information see Simulating on a Remote Machine.

Note

Communication between machines in remote analysis and distributed analysis can drastically affect performance. Use of a high-speed network system, like Gigabit or Infiniband, is recommended for optimal performance.

Note

(a) The machines entered in the Distributed Analysis Machines dialog are used for distributed solving. The machine that is set up in the Analysis tab is not considered.

(b) All the machines in this list are automatically selected for execution. Once the machines are entered into the list they will all be used for parallel launches.

(c) The machines entered in this dialog can be both local and remote. A machine can be entered into the list multiple times (for example, when there is a powerful server that can run multiple simulations).

(d) The number of parallel simulations that will be used for a parallel analysis is equal to the number of entries in the list of the Distributed Analysis Machines dialog.

(e) The limiting factor is the licenses you have. For example, if you have four licenses and ten machines in the list, and then start a parametric run with 10 iterations, Designer attempts to start ten simulation engines, but six of them will fail for lack of a license. The six iterations that have failed to get an engine will then be balanced and queued to the four engines that are running. Similarly, if there are ten iterations and five machines in the list, five engines will be started, one in each of the machines in the list. The remaining five iterations will be queued to run after the iterations that are already running complete analysis.

The distributed analysis uses an internal queuing system that is enabled whether or not the Queue all simulations option is selected on the Analysis Options tab. When both Queue all simulations and Distributed Analysis are in effect, the multiple simulations are queued, then run in parallel on the distributed machines as noted above.

HFSS视频教程 ADS视频教程 CST视频教程 Ansoft Designer 中文教程


Copyright © 2006 - 2013 微波EDA网, All Rights Reserved 业务联系：mweda@163.com