Local Network farm Core and Thread

[[Deleted]]

I am using my PC (8core/16thread) for backtest optimization and MT5 uses 16 threads. Then I setup another PC which as 4core/8threads as local network farm server. When connected and run optimization, only 4cores are used, instead of 8 thread?

Is it for remote farm server, only cores are used and thread?

[Sergey Golubev]

physical cores are used only in case of cloud and agents:

"Due to the apparent lack of memory with an excessive number of agents and a decrease in the speed of calculations on hyper-threading cores, we decided to limit ourselves to only physical cores when working in cloud" - post #60

[[Deleted]]

Sergey Golubev:

physical cores are used only in case of cloud and agents:

"Due to the apparent lack of memory with an excessive number of agents and a decrease in the speed of calculations on hyper-threading cores, we decided to limit ourselves to only physical cores when working in cloud" - post #60

I saw somewhere the recommendation is 2GB of memory per core.

These days 32GB per stick of memory is available and for a motherboard of 16 slots, it can hold up to 256GB of memory and up to 128 thread/core. Will MT5 consider to allow thread as well? Making an option in Stratgey Test Manager to detect amount of memory and if sufficient, allow thread as well, intended for personal use of the server and not intended to provide cloud service via MQL5 platform.

[Sergey Golubev]

firecard:

I saw somewhere the recommendation is 2GB of memory per core.

These days 32GB per stick of memory is available and for a motherboard of 16 slots, it can hold up to 256GB of memory and up to 128 thread/core. Will MT5 consider to allow thread as well? Making an option in Stratgey Test Manager to detect amount of memory and if sufficient, allow thread as well, intended for personal use of the server and not intended to provide cloud service via MQL5 platform.

I think - this limitation related to the physical cores will remain in the near future. But we will have the big improvement soon
This is the reply from MetaQuotes (machine translation to the English):

Forum on trading, automated trading systems and testing trading strategies

Cores were disappeared after updating the tester.

Renat Fatkhullin , 2020.07.25 18:20

In cloud, we only allow using the physical cores.

Statistics showed that many tasks in the cloud fail due to lack of resources. Very expensive experts went.

Over the past month, we have dramatically improved the MQL5 Cloud Network in terms of speed and responsiveness.

When we move to a new tester that will use one process for all cores instead of separate processes for each core, we will drastically reduce the need for resources due to consolidated caches and will be able to flexibly manage the load on all cores.

Here I described the plans in more detail: https://www.mql5.com/ru/forum/340770/page7#comment_16401357

Forum on trading, automated trading systems and testing trading strategies

New version of the MetaTrader 5 platform build 2450: "Subscriptions" service, interface improvements and convenient functions in MetaEditor

Renat Fatkhullin , 05/20/2016 12:35 PM

Yes, the tester will be rewritten/reprogrammed, as the current one is choked with data and an unsuccessful scaling scheme.

Each agent will work as a microcloud with N cores to keep one copy of the data for N threads. Since the time has come for the NUMA architecture among the masses, each agent will clearly live in its own NUMA node to avoid the penalties for accessing memory outside of its node.

This will significantly reduce memory requirements, reduce disk activity and increase the speed of optimization.

In this case, the main role will always be played by the volume of RAM, which allows you to keep more caches and quickly issue them.

The goal is to screen thousands of stocks in bulk, where we are heading with a new initiative to launch detailed stock data subscriptions.

[[Deleted]]

Sergey Golubev:

I think - this limitation related to the physical cores will remain in the near future. But we will have the big improvement soon
This is the reply from MetaQuotes (machine translation to the English):

Hello,

Thank for the reply.

Cloud here also refers to Local Agent Farm, right?

[[Deleted]]

Could someone comfirm that MT5 on Windows 10 Pro is able to make use of all cores of a 64 cores CPU (Ryzen™ Threadripper™ 3990X) or  Dual CPU (Intel XEON E5-2696v4) on 1 motherboard which give 44 cores in total for optimization of a EA?

[Euronis.com]

Hello to all serious MT5 Algo programmers and Optimisation teams.

We use custom symbols and finally in the 2734 build are able to use local agents to optimize default and custom symbols.

This screen shot shows it working

https://euronis.com/mt5-local-network-farm-v2734/

[Euronis.com]

firewind:
Could someone comfirm that MT5 on Windows 10 Pro is able to make use of all cores of a 64 cores CPU (Ryzen™ Threadripper™ 3990X) or  Dual CPU (Intel XEON E5-2696v4) on 1 motherboard which give 44 cores in total for optimization of a EA?

Which motherboard?

ryzen boards can handle it but you could run out of RAM depending on your symbols.

cheaper way to do is to build a local farm of PC's

[Markely Camara Leal]

I use Xeon 2420 on x79 MB, and they use as 16 cores. 8 cores x 2 threads...

[Adam L]

Markely Camara Leal #:
I use Xeon 2420 on x79 MB, and they use as 16 cores. 8 cores x 2 threads...

Ya, seems they aren't supporting a lot of the functionality they used to for local network farms... 8 threads on a quad core CPU work fine if there's plenty of RAM / HDD memory. Looks like whomever is driving change on MT5 strategy tester is more cloud-focused. Which is a shame as Local farms are reliable, private and more than enough for developing stable trading logic.

[[Deleted]]

They can monetized from cloud test, I guess that why they are keen on developing this further