3 <object id="TOC" data="graphical-toc.svg" type="image/svg+xml"></object>
5 window.onload=function() { // Wait for the SVG to be loaded before changing it
6 var elem=document.querySelector("#TOC").contentDocument.getElementById("ModelBox")
7 elem.style="opacity:0.93999999;fill:#ff0000;fill-opacity:0.1;stroke:#000000;stroke-width:0.35277778;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-dashoffset:0;stroke-opacity:1";
18 This page focuses on the **performance models** that compute the duration of :ref:`every activities <S4U_main_concepts>`
19 in the simulator depending on the platform characteristics and on the other activities that are currently sharing the
20 resources. If you look for other kind of models (such as routing models or compute model), please refer to :ref:`the
21 bottom of this page <models_other>`.
26 The main objective of SimGrid is to provide timing information for three following kind of resources: network, CPU,
29 The **network models** have been improved and regularly assessed for almost 20 years. It should be possible to get
30 accurate predictions once you properly :ref:`calibrate the models for your settings<models_calibration>`. As detailed
31 in the next sections, SimGrid provides several network models. Two plugins can also be used to compute the network
32 energy consumption: One for the :ref:`wired networks<plugin_link_energy>`, and another one for the :ref:`Wi-Fi networks
33 <plugin_link_energy>`. Some users find :ref:`TCP simulated performance counter-intuitive<understanding_lv08>` at first
34 in SimGrid, sometimes because of a misunderstanding of the TCP behavior in real networks.
36 The **computing models** are less developed in SimGrid. Through the S4U interface, the user specifies the amount of
37 computational work (expressed in FLOPs, for floating point operations) that each computation "consumes", and the model
38 simply divides this amount by the host's FLOP rate to compute the duration of this execution. In SMPI, the user code
39 is automatically timed, and the :ref:`computing speed<cfg=smpi/host-speed>` of the host machine is used to evaluate
40 the corresponding amount of FLOPs. This model should be sufficient for most users, even though assuming a constant FLOP
41 rate for each machine remains a crude simplification. In reality, the flops rate varies because of I/O, memory, and
42 cache effects. It is somehow possible to :ref:`overcome this simplification<cfg=smpi/comp-adjustment-file>`, but the
43 required calibration process is rather intricate and not documented yet (feel free to
44 :ref:`contact the community<community>` on need).
45 In the future, more advanced models may be added but the existing model proved good enough for all experiments done on
46 distributed applications during the last two decades. The CPU energy consumption can be computed with the
47 :ref:`relevant plugin<plugin_host_energy>`.
49 The **disk models** of SimGrid are more recent than those for the network and computing resources, but they should
50 still be correct for most users. `Studies have shown <https://hal.inria.fr/hal-01197128>`_ that they are sensitive
51 under some conditions, and a :ref:`calibration process<howto_disk>` is provided. As usual, you probably want to
52 double-check their predictions through an appropriate validation campaign.
57 SimGrid aims at the sweet spot between accuracy and simulation speed. About accuracy, our goal is to report correct
58 performance trends when comparing competing designs with a minimal burden on the user, while allowing power users to
59 fine tune the simulation models for predictions that are within 5% or less of the results on real machines. For
60 example, we determined the `speedup achieved by the Tibidabo ARM-based cluster <http://hal.inria.fr/hal-00919507>`_
61 before it was even built. About simulation speed, the tool must be fast and scalable enough to study modern IT systems
62 at scale. SimGrid was for example used to simulate `a Chord ring involving millions of actors
63 <https://hal.inria.fr/inria-00602216>`_ (even though that has not really been more instructive than smaller scale
64 simulations for this protocol), or `a qualification run at full-scale of the Stampede supercomputer
65 <https://hal.inria.fr/hal-02096571>`_.
67 Most of our models are based on a linear max-min solver (LMM), as depicted below. The actors' activities are
68 represented by actions in the simulation kernel, accounting for both the initial amount of work of the corresponding
69 activity (in FLOPs for computing activities or bytes for networking and disk activities), and the currently remaining
70 amount of work to process.
72 At each simulation step, the instantaneous computing and communicating capacity of each action is computed according
73 to the model. A set of constraints is used to express for example that the accumulated instantaneous consumption of a
74 given resource by a set actions must remain smaller than the nominal capacity speed of that resource. In the example
75 below, it is stated that the speed :math:`\varrho_1` of activity 1 plus the speed :math:`\varrho_n`
76 of activity :math:`n` must remain smaller than the capacity :math:`C_A` of the corresponding host A.
78 .. image:: img/lmm-overview.svg
80 There are obviously many valuations of :math:`\varrho_1, \ldots{}, \varrho_n` that respect such as set of constraints.
81 SimGrid usually computes the instantaneous speeds according to a Max-Min objective function, that is maximizing the
82 minimum over all :math:`\varrho_i`. The coefficients associated to each variable in the inequalities are used to model
83 some performance effects, such as the fact that TCP tends to favor communications with small RTTs. These coefficients
84 are computed from both hard-coded values and :ref:`latency and bandwidth factors<cfg=network/latency-factor>` (more
85 details on network performance modeling is given in the next section).
87 Once the instantaneous speeds are computed, the simulation kernel determines what is the earliest terminating action
88 from their respective speeds and remaining amounts of work. The simulated time is then updated along with the values
89 in the LMM. As some actions have nothing left to do, the corresponding activities thus terminate, which in turn
90 unblocks the corresponding actors that can further execute.
92 Most of the SimGrid models build upon the LMM solver, that they adapt and configure for their respective usage. For CPU
93 and disk activities, the LMM-based models are respectively named **Cas01** and **S19**. The existing network models are
94 described in the next section.
99 SimGrid provides several network performance models which compute the time taken by each communication in isolation.
100 **CM02** is the simplest one. It captures TCP windowing effects, but does not introduce any correction factors. This
101 model should be used if you prefer understandable results over realistic ones. **LV08** (the default model) uses
102 constant factors that are intended to capture common effects such as slow-start, the fact that TCP headers reduce the
103 *effective* bandwidth, or TCP's ACK messages. **SMPI** uses more advanced factors that also capture the MPI-specific
104 effects such as the switch between the eager vs. rendez-vous communication modes. You can :ref:`choose the
105 model<options_model_select>` on command line, and these models can be :ref:`further configured<options_model>`.
107 The LMM solver is then used as described above to compute the effect of contention on the communication time that is
108 computed by the TCP model. For sake of realism, the sharing on saturated links is not necessarily a fair sharing.
109 Instead, flows receive an amount of bandwidth inversely proportional to their round trip time.
111 Regardless of the used TCP model, the latency is paid beforehand. It is as if the communication only starts after a
112 little delay corresponding to the latency. During that time, the communication has no impact on the links (the other
113 communications are not slowed down, because there is no contention yet).
115 In addition to these LMM-based models, you can use the :ref:`ns-3 simulator as a network model <models_ns3>`. It is much
116 more detailed than the pure SimGrid models and thus slower, but it is easier to get more accurate results. Concerning
117 the speed, both simulators are linear in the size of their input, but ns-3 has a much larger input in case of large
118 steady communications. On the other hand, the SimGrid models must be carefully :ref:`calibrated <models_calibration>` if
119 accuracy is really important to your study, while ns-3 models are less demanding with that regard.
121 .. _understanding_cm02:
126 This is a simple model of TCP performance, where the sender stops sending packets when its TCP window is full. If the
127 acknowledgment packets are returned in time to the sender, the TCP window has no impact on the performance that then is
128 only limited by the link bandwidth. Otherwise, late acknowledgments will reduce the bandwidth.
130 SimGrid models this mechanism as follows: :math:`real\_BW = min(physical\_BW, \frac{TCP\_GAMMA}{2\times latency})` The used
131 bandwidth is either the physical bandwidth that is configured in the platform, or a value representing the bandwidth
132 limit due to late acknowledgments. This value is the maximal TCP window size (noted TCP Gamma in SimGrid) over the
133 round-trip time (i.e. twice the one-way latency). The default value of TCP Gamma is 4194304. This can be changed with
134 the :ref:`network/TCP-gamma <cfg=network/TCP-gamma>` configuration item.
136 If you want to disable this mechanism altogether (to model e.g. UDP or memory movements), you should set TCP-gamma
137 to 0. Otherwise, the time it takes to send 10 Gib of data over a 10 Gib/s link that is otherwise unused is computed as
138 follows. This is always given by :math:`latency + \frac{size}{bandwidth}`, but the bandwidth to use may be the physical
139 one (10Gb/s) or the one induced by the TCP window, depending on the latency.
141 - If the link latency is 0, the communication obviously takes one second.
142 - If the link latency is 0.00001s, :math:`\frac{gamma}{2\times lat}=209,715,200,000 \approx 209Gib/s` which is larger than the
143 physical bandwidth. So the physical bandwidth is used (you fully use the link) and the communication takes 1.00001s
144 - If the link latency is 0.001s, :math:`\frac{gamma}{2\times lat}=2,097,152,000 \approx 2Gib/s`, which is smalled than the
145 physical bandwidth. The communication thus fails to fully use the link, and takes about 4.77s.
146 - With a link latency of 0.1s, :math:`gamma/2\times lat \approx 21Mb/s`, so the communication takes about 476.84 + 0.1 seconds!
147 - More cases are tested and enforced by the test ``teshsuite/models/cm02-tcpgamma/cm02-tcpgamma.tesh``
149 For more details, please refer to "A Network Model for Simulation of Grid Application" by Henri Casanova and Loris
150 Marchal (published in 2002, thus the model name).
152 .. _understanding_lv08:
157 This model builds upon CM02 to model TCP windowing. It also introduces corrections factors for further realism. Lets
158 consider the following platform:
162 <host id="A" speed="1Gf" />
163 <host id="B" speed="1Gf" />
165 <link id="link1" latency="10ms" bandwidth="1Mbps" />
167 <route src="A" dst="B">
168 <link_ctn id="link1" />
171 If host `A` sends ``100kB`` (a hundred kilobytes) to host `B`, one can expect that this communication would take `0.81`
172 seconds to complete according to a simple latency-plus-size-divided-by-bandwidth model (0.01 + 8e5/1e6 = 0.81) since the
173 latency is small enough to ensure that the physical bandwidth is used (see the discussion on CM02 above). However, the
174 LV08 model is more complex to account for three phenomena that directly impact the simulation time:
176 - The size of a message at the application level (i.e., 100kB in this example) is not the size that is actually
177 transferred over the network. To mimic the fact that TCP and IP headers are added to each packet of the original
178 payload, the TCP model of SimGrid empirically considers that `only 97% of the nominal bandwidth` are available. In
179 other words, the size of your message is increased by a few percents, whatever this size be.
181 - In the real world, the TCP protocol is not able to fully exploit the bandwidth of a link from the emission of the
182 first packet. To reflect this `slow start` phenomenon, the latency declared in the platform file is multiplied by
183 `a factor of 13.01`. Here again, this is an empirically determined value that may not correspond to every TCP
184 implementations on every networks. It can be tuned when more realistic simulated times for the transfer of short
185 messages are needed though.
187 - When data is transferred from A to B, some TCP ACK messages travel in the opposite direction. To reflect the impact
188 of this `cross-traffic`, SimGrid simulates a flow from B to A that represents an additional bandwidth consumption
189 of `0.05%`. The route from B to A is implicitly declared in the platform file and uses the same link `link1` as if
190 the two hosts were connected through a communication bus. The bandwidth share allocated to a data transfer from A
191 to B is then the available bandwidth of `link1` (i.e., 97% of the nominal bandwidth of 1Mb/s) divided by 1.05
192 (i.e., the total consumption). This feature, activated by default, can be disabled by adding the
193 ``--cfg=network/crosstraffic:0`` flag to the command line.
195 As a consequence, the time to transfer 100kB from A to B as simulated by the default TCP model of SimGrid is not 0.81
198 .. code-block:: python
200 0.01 * 13.01 + 800000 / ((0.97 * 1e6) / 1.05) = 0.996079 seconds.
202 For more details, please refer to "Accuracy study and improvement of network simulation in the SimGrid framework" by
203 Arnaud Legrand and Pedro Velho.
208 This model is rather distinct from the other LMM models because it uses another objective function called *bottleneck*.
209 This is because this model is intended to be used for parallel tasks that are actions mixing flops and bytes while the
210 Max-Min objective function requires that all variables are expressed using the same unit. This is also why in reality,
211 we have one LMM system per resource kind in the simulation, but the idea remains similar.
218 In SimGrid, WiFi networks are modeled with WiFi zones, where a zone contains the access point of the WiFi network and
219 the hosts connected to it (called `stations` in the WiFi world). The network inside a WiFi zone is modeled by declaring
220 a single regular link with a specific attribute. This link is then added to the routes to and from the stations within
221 this WiFi zone. The main difference of WiFi networks is that their performance is not determined by some link bandwidth
222 and latency but by both the access point WiFi characteristics and the distance between that access point and a given
225 Such WiFi zones can be used with the LMM-based model or ns-3, and are supposed to behave similarly in both cases.
227 Declaring a WiFi zone
228 =====================
230 To declare a new WiFi network, simply declare a network zone with the ``WIFI`` routing attribute.
234 <zone id="SSID_1" routing="WIFI">
236 Inside this zone you must declare which host or router will be the access point of the WiFi network.
240 <prop id="access_point" value="alice"/>
242 Then simply declare the stations (hosts) and routers inside the WiFi network. Remember that one must have the same name
243 as the "access point" property.
247 <router id="alice" speed="1Gf"/>
248 <host id="STA0-0" speed="1Gf"/>
249 <host id="STA0-1" speed="1Gf"/>
251 Finally, close the WiFi zone.
257 The WiFi zone may be connected to another zone using a traditional link and a zoneRoute. Note that the connection between two
258 zones is always wired.
262 <link id="wireline" bandwidth="100Mbps" latency="2ms" sharing_policy="SHARED"/>
264 <zoneRoute src="SSID_1" dst="SSID_2" gw_src="alice" gw_dst="bob">
265 <link_ctn id="wireline"/>
268 WiFi network performance
269 ========================
271 The performance of a wifi network is controlled by the three following properties:
273 * ``mcs`` (`Modulation and Coding Scheme <https://en.wikipedia.org/wiki/Link_adaptation>`_)
274 is a property of the WiFi zone. Roughly speaking, it defines the speed at which the access point is exchanging data
275 with all the stations. It depends on the access point's model and configuration. Possible values for the MCS can be
276 found on Wikipedia for example.
277 |br| By default, ``mcs=3``.
278 * ``nss`` (Number of Spatial Streams, or `number of antennas <https://en.wikipedia.org/wiki/IEEE_802.11n-2009#Number_of_antennas>`_) is another property of the WiFi zone. It defines the amount of simultaneous data streams that the access
279 point can sustain. Not all values of MCS and NSS are valid nor compatible (cf. `802.11n standard <https://en.wikipedia.org/wiki/IEEE_802.11n-2009#Data_rates>`_).
280 |br| By default, ``nss=1``.
281 * ``wifi_distance`` is the distance from a station to the access point. Each station can have its own specific value.
282 It is thus a property of the stations declared inside the WiFi zone.
283 |br| By default, ``wifi_distance=10``.
285 Here is an example of a zone with non-default ``mcs`` and ``nss`` values.
289 <zone id="SSID_1" routing="WIFI">
290 <prop id="access_point" value="alice"/>
291 <prop id="mcs" value="2"/>
292 <prop id="nss" value="2"/>
296 Here is an example of setting the ``wifi_distance`` of a given station.
300 <host id="STA0-0" speed="1Gf">
301 <prop id="wifi_distance" value="37"/>
307 This simplistic network model is one of the few SimGrid network model that is not based on the LMM solver. In this
308 model, all communication take a constant time (one second by default). It provides the lowest level of realism, but is
309 marginally faster and much simpler to understand. This model may reveal interesting if you plan to study abstract
310 distributed algorithms such as leader election or causal broadcast.
314 ns-3 as a SimGrid model
315 ***********************
317 The **ns-3 based model** is the most accurate network model that you can get in SimGrid. It relies on the well-known
318 `ns-3 packet-level network simulator <http://www.nsnam.org>`_ to compute every timing information related to the network
319 transfers of your simulation. For instance, this may be used to investigate the validity of a simulation. Note that this
320 model is much slower than the LMM-based models, because ns-3 simulates the movement of every network packet involved in
321 every communication while SimGrid only recomputes the respective instantaneous speeds of the currently ongoing
322 communications when one communication starts or stops.
324 You need to install ns-3 and recompile SimGrid accordingly to use this model.
326 The SimGrid/ns-3 binding only contains features that are common to both systems. Not all ns-3 models are available from
327 SimGrid (only the TCP and WiFi ones are), while not all SimGrid platform files can be used in conjunction with ns-3
328 (routes must be of length 1). Also, the platform built in ns-3 from the SimGrid
329 description is very basic. Finally, communicating from a host to
330 itself is forbidden in ns-3, so every such communication completes
331 immediately upon startup.
334 Compiling the ns-3/SimGrid binding
335 ==================================
340 SimGrid requires ns-3 version 3.26 or higher, and you probably want the most
341 recent version of both SimGrid and ns-3. While the Debian package of SimGrid
342 does not have the ns-3 bindings activated, you can still use the packaged version
343 of ns-3 by grabbing the ``libns3-dev ns3`` packages. Alternatively, you can
344 install ns-3 from scratch (see the `ns-3 documentation <http://www.nsnam.org>`_).
346 Enabling ns-3 in SimGrid
347 ------------------------
349 SimGrid must be recompiled with the ``enable_ns3`` option activated in cmake.
350 Optionally, use ``NS3_HINT`` to tell cmake where ns3 is installed on
353 .. code-block:: console
355 $ cmake . -Denable_ns3=ON -DNS3_HINT=/opt/ns3 # or change the path if needed
357 By the end of the configuration, cmake reports whether ns-3 was found,
358 and this information is also available in ``include/simgrid/config.h``
359 If your local copy defines the variable ``SIMGRID_HAVE_NS3`` to 1, then ns-3
360 was correctly detected. Otherwise, explore ``CMakeFiles/CMakeOutput.log`` and
361 ``CMakeFiles/CMakeError.log`` to diagnose the problem.
363 Test that ns-3 was successfully integrated with the following command (executed from your SimGrid
364 build directory). It will run all SimGrid tests that are related to the ns-3
365 integration. If no test is run at all, then ns-3 is disabled in cmake.
367 .. code-block:: console
374 If you use a version of ns-3 that is not known to SimGrid yet, edit
375 ``tools/cmake/Modules/FindNS3.cmake`` in your SimGrid tree, according to the
376 comments on top of this file. Conversely, if something goes wrong with an old
377 version of either SimGrid or ns-3, try upgrading everything.
379 Note that there is a known bug with the version 3.31 of ns3 when it is built with
380 MPI support, like it is with the libns3-dev package in Debian 11 « Bullseye ».
381 A simple workaround is to edit the file
382 ``/usr/include/ns3.31/ns3/point-to-point-helper.h`` to remove the ``#ifdef NS3_MPI``
383 include guard. This can be achieved with the following command (as root):
385 .. code-block:: console
387 # sed -i '/^#ifdef NS3_MPI/,+2s,^#,//&,' /usr/include/ns3.31/ns3/point-to-point-helper.h
391 Using ns-3 from SimGrid
392 =======================
394 Platform files compatibility
395 ----------------------------
397 Any route longer than one will be ignored when using ns-3. They are
398 harmless, but you still need to connect your hosts using one-hop routes.
399 The best solution is to add routers to split your route. Here is an
400 example of an invalid platform:
404 <?xml version='1.0'?>
405 <!DOCTYPE platform SYSTEM "https://simgrid.org/simgrid.dtd">
406 <platform version="4.1">
407 <zone id="zone0" routing="Floyd">
408 <host id="alice" speed="1Gf" />
409 <host id="bob" speed="1Gf" />
411 <link id="l1" bandwidth="1Mbps" latency="5ms" />
412 <link id="l2" bandwidth="1Mbps" latency="5ms" />
414 <route src="alice" dst="bob">
415 <link_ctn id="l1"/> <!-- !!!! IGNORED WHEN USED WITH ns-3 !!!! -->
416 <link_ctn id="l2"/> <!-- !!!! ROUTES MUST CONTAIN ONE LINK ONLY !!!! -->
421 This can be reformulated as follows to make it usable with the ns-3 binding.
422 There is no direct connection from alice to bob, but that's OK because ns-3
423 automatically routes from point to point (using
424 ``ns3::Ipv4GlobalRoutingHelper::PopulateRoutingTables``).
428 <?xml version='1.0'?>
429 <!DOCTYPE platform SYSTEM "https://simgrid.org/simgrid.dtd">
430 <platform version="4.1">
431 <zone id="zone0" routing="Full">
432 <host id="alice" speed="1Gf" />
433 <host id="bob" speed="1Gf" />
435 <router id="r1" /> <!-- routers are compute-less hosts -->
437 <link id="l1" bandwidth="1Mbps" latency="5ms"/>
438 <link id="l2" bandwidth="1Mbps" latency="5ms"/>
440 <route src="alice" dst="r1">
444 <route src="r1" dst="bob">
450 Once your platform is OK, just change the :ref:`network/model
451 <options_model_select>` configuration option to `ns-3` as follows. The other
452 options can be used as usual.
454 .. code-block:: console
456 $ ./network-ns3 --cfg=network/model:ns-3 (other parameters)
458 Many other files from the ``examples/platform`` directory are usable with the
459 ns-3 model, such as `examples/platforms/dogbone.xml <https://framagit.org/simgrid/simgrid/tree/master/examples/platforms/dogbone.xml>`_.
460 Check the file `examples/cpp/network-ns3/network-ns3.tesh <https://framagit.org/simgrid/simgrid/tree/master/examples/cpp/network-ns3/network-ns3.tesh>`_
461 to see which ones are used in our regression tests.
463 Alternatively, you can manually modify the ns-3 settings by retrieving
464 the ns-3 node from any given host with the
465 :cpp:func:`simgrid::get_ns3node_from_sghost` function (defined in
466 ``simgrid/plugins/ns3.hpp``).
468 .. doxygenfunction:: simgrid::get_ns3node_from_sghost
472 It is possible to define a fixed or random seed to the ns3 random number generator using the config tag.
476 <?xml version='1.0'?><!DOCTYPE platform SYSTEM "https://simgrid.org/simgrid.dtd">
477 <platform version="4.1">
479 <prop id = "network/model" value = "ns-3" />
480 <prop id = "ns3/seed" value = "time" />
485 The first property defines that this platform will be used with the ns3 model.
486 The second property defines the seed that will be used. Defined to ``time``,
487 it will use a random seed, defined to a number it will use this number as
493 A ns-3 platform is automatically created from the provided SimGrid
494 platform. However, there are some known caveats:
496 * The default values (e.g., TCP parameters) are the ns-3 default values.
497 * ns-3 networks are routed using the shortest path algorithm, using ``ns3::Ipv4GlobalRoutingHelper::PopulateRoutingTables``.
498 * End hosts cannot have more than one interface card. So, your SimGrid hosts
499 should be connected to the platform through only one link. Otherwise, your
500 SimGrid host will be considered as a router (FIXME: is it still true?).
502 Our goal is to keep the ns-3 plugin of SimGrid as easy (and hopefully readable)
503 as possible. If the current state does not fit your needs, you should modify
504 this plugin, and/or create your own plugin from the existing one. If you come up
505 with interesting improvements, please contribute them back.
510 If your simulation hangs in a communication, this is probably because one host
511 is sending data that is not routable in your platform. Make sure that you only
512 use routes of length 1, and that any host is connected to the platform.
513 Arguably, SimGrid could detect this situation and report it, but unfortunately,
514 this still has to be done.
519 `FMI <https://fmi-standard.org/>`_ is a standard to exchange models between simulators. If you want to plug such a model
520 into SimGrid, you need the `SimGrid-FMI external plugin <https://framagit.org/simgrid/simgrid-FMI>`_.
521 There is a specific `documentation <https://simgrid.frama.io/simgrid-FMI/index.html>`_ available for the plugin.
522 This was used to accurately study a *Smart grid* through co-simulation: `PandaPower <http://www.pandapower.org/>`_ was
523 used to simulate the power grid, `ns-3 <https://nsnam.org/>`_ was used to simulate the communication network while SimGrid was
524 used to simulate the IT infrastructure. Please also refer to the `relevant publication <https://hal.science/hal-03217562>`_
532 As for any simulator, models are very important components of the SimGrid toolkit. Several kind of models are used in
533 SimGrid beyond the performance models described above:
535 The **routing models** constitute advanced elements of the platform description. This description naturally entails
536 :ref:`components<platform>` that are very related to the performance models. For instance, determining the execution
537 time of a task obviously depends on the characteristics of the machine that executes this task. Furthermore, networking
538 zones can be interconnected to compose larger platforms `in a scalable way <http://hal.inria.fr/hal-00650233/>`_. Each
539 of these zones can be given a specific :ref:`routing model<platform_routing>` that efficiently computes the list of
540 links forming a network path between two given hosts.
542 The model checker uses an abstraction of the performance simulations. Mc SimGrid explores every causally possible
543 execution paths of the application, completely abstracting the performance away. The simulated time is not even
544 computed in this mode! The abstraction involved in this process also models the mutual impacts among actions, to not
545 re-explore histories that only differ by the order of independent and unrelated actions. As with the rest of the model
546 checker, these models are unfortunately still to be documented properly.