When commercial pilots train for disasters it's not with Microsoft Flight
Simulator, but something more like this realistic simulator.
You too can avoid outages like the recent Southwest grounding by training
your network operators to handle unforeseen scenarios with the industry-
leading MIMIC Simulator.
Create network labs that approximate your production network and play
out complex scenarios on demand to implement / test / train network
management policies. Train your IT staff with all the latest devices and
various network environments. That way they are prepared when a disaster
happens and resolve it quickly.
Thursday, July 21, 2016
Friday, July 8, 2016
Large topologies with MIMIC for development, testing, sales demos
Would you prefer a topology like this
With MIMIC Simulator Enhanced Topology Wizard, you can create many
different synthetic, large-scale, verifiable networks in minutes to
drive development, testing, sales demonstrations of your configuration
management application that visualizes topologies.
For example, here are different 500-node topologies:
Too symmetric? Spend a couple of minutes more and create something like this:
or like this
to verify / demonstrate your network management application?
With MIMIC Simulator Enhanced Topology Wizard, you can create many
different synthetic, large-scale, verifiable networks in minutes to
drive development, testing, sales demonstrations of your configuration
management application that visualizes topologies.
For example, here are different 500-node topologies:
Too symmetric? Spend a couple of minutes more and create something like this:
Thursday, July 7, 2016
Long-term latency test with MIMIC MQTT Simulator
If your large-scale IoT application has latency requirements, you want
to do long-term testing to ensure your environment satisfies those
requirements.
We ran the end-to-end latency test as shown in this published video
with MIMIC MQTT Simulator against a third-party MQTT broker setup
overnight, and graphed the latency reported by our subscriber client
action script with Grafana.
The setup simulates 3000 MQTT clients, of which one is the subscriber
that measures round-trip delay of published messages to unique topics.
This tests a common fan-out scenario.
We did not expect to find anything interesting. The attached picture
shows a good steady state, with average and max latency times below
and around 500 msec (respectively) ... most of the time.
But, there are 2 instances where max latency times are above 1 second.
If your IoT application requirement is latency times below 1 second,
you'd be worried, and would need to start investigating. Without this
test you would never know about any long-term problems in your setup.
to do long-term testing to ensure your environment satisfies those
requirements.
We ran the end-to-end latency test as shown in this published video
with MIMIC MQTT Simulator against a third-party MQTT broker setup
overnight, and graphed the latency reported by our subscriber client
action script with Grafana.
The setup simulates 3000 MQTT clients, of which one is the subscriber
that measures round-trip delay of published messages to unique topics.
This tests a common fan-out scenario.
We did not expect to find anything interesting. The attached picture
shows a good steady state, with average and max latency times below
and around 500 msec (respectively) ... most of the time.
But, there are 2 instances where max latency times are above 1 second.
If your IoT application requirement is latency times below 1 second,
you'd be worried, and would need to start investigating. Without this
test you would never know about any long-term problems in your setup.
Wednesday, July 6, 2016
MQTT Simulator enables large-scale Azure IoT testing
We have recently simulated many sensors with MIMIC MQTT Simulator
publishing arbitrary messages to a subscriber client connected to
Azure IoT Hub as a proof of concept of large-scale testing of IoT application
deployment using this platform.
publishing arbitrary messages to a subscriber client connected to
Azure IoT Hub as a proof of concept of large-scale testing of IoT application
deployment using this platform.
Friday, June 24, 2016
MIMIC MQTT Simulator interoperates with major brokers
In addition to our prior published reports, we recently simulated
thousands of sensors to 3 more brokers
VerneMQ
RabbitMQ
EMQTT
with thousands of arbitrary messages per second published.
Regardless which broker you select, you can perform prototyping, scalability,
robustness and security testing of your IoT application with
MIMIC MQTT Simulator prior to deployment.
thousands of sensors to 3 more brokers
VerneMQ
RabbitMQ
EMQTT
with thousands of arbitrary messages per second published.
Regardless which broker you select, you can perform prototyping, scalability,
robustness and security testing of your IoT application with
MIMIC MQTT Simulator prior to deployment.
Friday, June 17, 2016
How is realistic MQTT testing with MIMIC different from simplistic testing?
Most simple test apps use a for loop to create connections to a MQTT broker,
using only one single IP address for a large number of client sessions.
By using one IP address per session with MIMIC MQTT Simulator just like in
real world scenarios (when do real sensors share IP addresses?) we
uncovered a flaw in third-party MQTT brokers.
They would fail after a small number of simulated sensors (typically 1000
connections).
Only after adjusting the ARP cache parameters on the broker systems did
we overcome this problem.
MIMIC allows you to create realistic IoT pre-deployment testing scenarios
using IPv4 or IPv6 addressing, just like in the real world.
using only one single IP address for a large number of client sessions.
By using one IP address per session with MIMIC MQTT Simulator just like in
real world scenarios (when do real sensors share IP addresses?) we
uncovered a flaw in third-party MQTT brokers.
They would fail after a small number of simulated sensors (typically 1000
connections).
Only after adjusting the ARP cache parameters on the broker systems did
we overcome this problem.
MIMIC allows you to create realistic IoT pre-deployment testing scenarios
using IPv4 or IPv6 addressing, just like in the real world.
Friday, June 3, 2016
MIMIC MQTT Simulator publishing synthetic load to HiveMQ
MIMIC MQTT Simulator is simulating 40,000 connected sensors.
The HiveMQ MQTT Broker connected with them as if they are physical
sensors. As shown in the graphs, MIMIC is generating messages at the
rate of around 4,000 messages / second.
In this scenario it has been running for many hours.
The 40001st connection is a third-party subscriber client. This shows
that MIMIC simulated sensors are no different than the real one.
MIMIC allows you to stage precise pre-deployment tests for HiveMQ and
other IoT applications to make sure they handle the expected load and
provide the needed performance. In addition, you can easily customize
message payloads of simulated sensors for rapid prototyping, and create
pathological scenarios to minimize risks.
The HiveMQ MQTT Broker connected with them as if they are physical
sensors. As shown in the graphs, MIMIC is generating messages at the
rate of around 4,000 messages / second.
In this scenario it has been running for many hours.
The 40001st connection is a third-party subscriber client. This shows
that MIMIC simulated sensors are no different than the real one.
MIMIC allows you to stage precise pre-deployment tests for HiveMQ and
other IoT applications to make sure they handle the expected load and
provide the needed performance. In addition, you can easily customize
message payloads of simulated sensors for rapid prototyping, and create
pathological scenarios to minimize risks.
Subscribe to:
Posts (Atom)