RPA high load
It's very important to understand and use parallel execution techniques which the WorkFusion infrastructure provides.
The Selenium infrastructure topology looks as below:

For example, you have multiple Nodes which could be pushed with Selenium code and work in parallel.
The solution is as follows:
Design your BP.
Put your RPA transaction into a single Bot Task. A transaction here means any number of linked together selenium blocks within selenium-flow. In those blocks, applications interchange with data, and their session is held to continue in the following selenium block.
For MT with an RPA transaction, configure Source. It should have Thread Count = N-1 where N is number of Selenium Nodes you have in the infrastructure.
After you configure as described above, you can launch and connect via RDP into RPA Agent Windows box to make sure it executes in parallel.
As an example, BP is simply on MT with RPA code:

Configure Bot Source via UI:


GMail and Google search RPA steps source
<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://web-harvest.sourceforge.net/schema/1.0/config" scriptlang="groovy">
<var-def name="session_uuid">
<template>${UUID.randomUUID().toString()}</template>
</var-def>
<robotics-flow>
<robot name="robotDriver" driver="internet explorer" close-on-completion="false" start-in-private="true">
<script><![CDATA[
timeouts(15000)
open("http://gmail.com")
def emailControl = $(byId("Email"))
emailControl.click()
emailControl.clear()
emailControl.sendKeys("tuk.tuk.rpa@gmail.com")
$(byId("next")).click()
def passwordControl = $(byId("Passwd"))
passwordControl.click()
passwordControl.clear()
passwordControl.sendKeys("rpapassword")
$(byId("signIn")).click()
sleep(2000)
List elem = $$(byXpath("//a[contains(.,'No, not interested')]"))
if (elem.size()>0) { elem.get(0).click() }
]]></script>
</robot>
<robot name="robotDriverOne" driver="internet explorer" close-on-completion="true" start-in-private="true">
<script></script>
</robot>
<var-def name="xmlValue">
<html-to-xml>
<template>${xmlValue}</template>
</html-to-xml>
</var-def>
<var-def name="last_checked">
<script return="result"></script>
</var-def>
<var-def name="currentDate">
<script return="result"></script>
</var-def>
<var-def name="s3Links">
<s3 bucket="your-bucket">
<s3-put-public path="degs/search_res_${currentDate}/${last_checked}.html" content="${xmlValue}" content-type="text/html" content-disposition="inline" />
</s3>
</var-def>
<script><![CDATA[
$(byXpath("//div[text()='COMPOSE']")).click()
def to = $(byName("to"))
to.click()
sleep(2000)
to.sendKeys("tuk.tuk.rpa@gmail.com")
sleep(1000)
def subject = $(byName("subjectbox"))
subject.click()
sleep(2000)
subject.sendKeys("Test message from ROBOT (MT: GMail and Google Search RPA steps)")
sleep(2000)
def body = $(byXpath("//div[@aria-label='Message Body']"))
body.click()
sleep(2000)
body.sendKeys("Link extracted : " + s3Links.toString())
sleep(2000)
$(byXpath("//div[text()='Send']")).click()
driver().close()
driver().quit()
]]></script>
</robotics-flow>
<export include-original-data="false">
<single-column name="session_uuid" value='${session_uuid}'/>
</export>
</config>