Can you help do build this model ?

Question

question

Дмитрий asked Dec 22, '20 Ben Wilson commented Jan 6, '21

Can you help do build this model ?

In the digital information transmission system

broadcasting is transmitted digitally. Language packs

transmitted through two transit channels, buffered in

drives in front of each channel. Software packet transfer time

the channel is 5 ms. Packets arrive after 6±3 ms. Packages that

they were transmitted for more than 10 MS, and the systems are destroyed at the output, so

that their appearance in the decoder will significantly reduce the quality of transmitted information

broadcasting. Destruction of more than 30% of packages is unacceptable. When

reaching this level, the system uses resources to speed up

transmission up to 4 ms per channel. When the level is lowered to an acceptable level

resources are being disabled.

Determine the frequency of packet destruction and frequency

enabling a resource.

Software Version:

FlexSim 20.1.0

flexsim 20.1.0 flexisim transfer protocol broadcast processing data

· 3

Joerg Vogel commented · Dec 23, 2020 at 07:28 AM

Hi Dimitrii, I need some more information to answer your request.

Do you have a frequency of broadcasts or a statistical distribution or them? this is related to later point 3.
Any operation lasting longer than 10ms will result in a destroyed broadcast. Is this correct? Transfer Time + buffering time in two drives must be smaller than 10 ms. The processing time is 6ms +/- 3ms normal distributed. Only one broadcast can be processed.
So you need a self-controlled model that turns resources on only if the destruction level rises above 30%? This is independent on the condition of the frequency of broadcasts.

The result might be a relation of frequency of system input to frequency of destroyed broadcasts.

0 ·

Joerg Vogel commented · Dec 23, 2020 at 08:10 AM

Dimtrii, can you ask a bit more descriptive question? Something like ”optimize broadcast packages” or “process information data”. Thx.

0 ·

Ben Wilson ♦♦ commented · Jan 06, 2021 at 12:11 PM

Hi @Дмитрий , was Jörg Vogel's answer helpful? If so, please click the red "Accept" button on their answer. Or if you still have questions, add a comment and we'll continue the conversation.

If we haven't heard back from you within 3 business days we'll auto-accept an answer, but you can always unaccept and comment back to reopen your question.

0 ·

Answer 1 · 2020-12-23T07:51:13Z

Joerg Vogel answered Dec 23, '20

If my assumptions are correct, you drag a processor (a) into your model with a label driven process time of 5ms. Then the output is connected to two queues. They are your drives. The items are sent to them by round robin property. The queue sends items to the output port number of the following processor. This processor has got a mean process time of 6 ms with a normal distributed deviation of 3ms. If stored items in the queues are staying longer than 10ms in system they are released to a sink, before they enter the processor.

An approach to release items to another output port is possible by the function releaseitem. Whenever an item enters a queue a delayed message is sent from the item to the queue (10ms - current transfer time) from the OnEntry trigger.

0.01 - (Model.time - getentrytime(item))

If the message is received in the OnMessage trigger and the msgsendingobject is still in the queue then the item is released to output port connected to a sink.

source code OnMessage trigger:

treenode item = msgsendingobject();
if(item.up == current) relaseitem(item,2);// port 2 goes to sink.

You put items entering and being processed in relation. If this relation drops below 70% you set the label value of processor (a) to 4ms.

If the successful processing rate rises again above 70% you adjust the processing time label of processor (a) to 5ms again.

You set the input arrival time of your model by label parameters in the source and you do some experiments to get the answer at which input rate the system is stable.

You can reset the ratio from time to time back to 1 just to change the dynamic of the response.

question