1 /*! @page platform Describing the virtual platform
4 @section pf_res Resource description
6 @subsection pf_res_computing Computing Resources
9 @subsubsection pf_tag_cluster <cluster>
11 ``<cluster />`` represents a machine-cluster. It is most commonly used
12 when one wants to define many hosts and a network quickly. Technically,
13 ``cluster`` is a meta-tag: <b>from the inner SimGrid point of
14 view, a cluster is a network zone where some optimized routing is defined</b>.
15 The default inner organization of the cluster is as follow:
21 ____________|__________|_____________ backbone
23 l0| l1| l2| l97| l96 | | l99
29 Here, a set of <b>host</b>s is defined. Each of them has a <b>link</b>
30 to a central backbone (backbone is a link itself, as a link can
31 be used to represent a switch, see the switch / link section
32 below for more details about it). A <b>router</b> allows to connect a
33 <b>cluster</b> to the outside world. Internally,
34 SimGrid treats a cluster as a network zone containing all hosts: the router is the default
35 gateway for the cluster.
37 There is an alternative organization, which is as follows:
51 The principle is the same, except that there is no backbone. This representation
52 can be obtained easily: just do not set the bb_* attributes.
55 Attribute name | Mandatory | Values | Description
56 --------------- | --------- | ------ | -----------
57 id | yes | string | The identifier of the cluster. Facilitates referring to this cluster.
58 prefix | yes | string | Each node of the cluster has to have a name. This name will be prefixed with this prefix.
59 suffix | yes | string | Each node of the cluster will be suffixed with this suffix
60 radical | yes | string | Regexp used to generate cluster nodes name. Syntax: "10-20" will give you 11 machines numbered from 10 to 20, "10-20;2" will give you 12 machines, one with the number 2, others numbered as before. The produced number is concatenated between prefix and suffix to form machine names.
61 speed | yes | int | Same as the ``speed`` attribute of the ``@<host@>`` tag.
62 core | no | int (default: 1) | Same as the ``core`` attribute of the ``@<host@>`` tag.
63 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the @ref pf_tag_link "link section" for syntax/details.
64 lat | yes | int | Latency for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
65 sharing_policy | no | string | Sharing policy for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
66 bb_bw | no | int | Bandwidth for backbone (if any). See <b>link</b> section for syntax/details. If bb_bw and bb_lat (see below) attributes are omitted, no backbone is created (alternative cluster architecture <b>described before</b>).
67 bb_lat | no | int | Latency for backbone (if any). See <b>link</b> section for syntax/details. If bb_lat and bb_bw (see above) attributes are omitted, no backbone is created (alternative cluster architecture <b>described before</b>).
68 bb_sharing_policy | no | string | Sharing policy for the backbone (if any). See <b>link</b> section for syntax/details.
69 limiter_link | no | int | Bandwidth for limiter link (if any). This adds a specific link for each node, to set the maximum bandwidth reached when communicating in both directions at the same time. In theory this value should be 2*bw for splitduplex links, but in reality this might be less. This value will depend heavily on the communication model, and on the cluster's hardware, so no default value can be set, this has to be measured. More details can be obtained in <a href="https://hal.inria.fr/hal-00919507/"> "Toward Better Simulation of MPI Applications on Ethernet/TCP Networks"</a>
70 loopback_bw | no | int | Bandwidth for loopback (if any). See <b>link</b> section for syntax/details. If loopback_bw and loopback_lat (see below) attributes are omitted, no loopback link is created and all intra-node communication will use the main network link of the node. Loopback link is a @ref pf_sharing_policy_fatpipe "@b FATPIPE".
71 loopback_lat | no | int | Latency for loopback (if any). See <b>link</b> section for syntax/details. See loopback_bw for more info.
72 topology | no | FLAT@|TORUS@|FAT_TREE@|DRAGONFLY (default: FLAT) | Network topology to use. SimGrid currently supports FLAT (with or without backbone, as described before), <a href="http://en.wikipedia.org/wiki/Torus_interconnect">TORUS </a>, FAT_TREE, and DRAGONFLY attributes for this tag.
73 topo_parameters | no | string | Specific parameters to pass for the topology defined in the topology tag. For torus networks, comma-separated list of the number of nodes in each dimension of the torus. Please refer to the specific documentation for @ref simgrid::kernel::routing::FatTreeZone "FatTree NetZone", @ref simgrid::kernel::routing::DragonflyZone "Dragonfly NetZone".
76 the router name is defined as the resulting String in the following
80 router_name = prefix + clusterId + "_router" + suffix;
84 #### Cluster example ####
86 Consider the following two (and independent) uses of the ``cluster`` tag:
89 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
90 speed="1e9" bw="125e6" lat="5E-5"/>
92 <cluster id="my_cluster_2" prefix="c-" suffix=".me" radical="0-99"
93 speed="1e9" bw="125e6" lat="5E-5"
94 bb_bw="2.25e9" bb_lat="5E-4"/>
97 The second example creates one router and 100 machines with the following names:
99 c-my_cluster_2_router.me
107 @subsubsection pf_cabinet <cabinet>
110 This tag is only available when the routing mode of the network zone
111 is set to ``Cluster``.
113 The ``<cabinet />`` tag is, like the @ref pf_tag_cluster "<cluster>" tag,
114 a meta-tag. This means that it is simply a shortcut for creating a set of (homogenous) hosts and links quickly;
115 unsurprisingly, this tag was introduced to setup cabinets in data centers quickly. Unlike
116 <cluster>, however, the <cabinet> assumes that you create the backbone
117 and routers yourself; see our examples below.
121 Attribute name | Mandatory | Values | Description
122 --------------- | --------- | ------ | -----------
123 id | yes | string | The identifier of the cabinet. Facilitates referring to this cluster.
124 prefix | yes | string | Each node of the cabinet has to have a name. This name will be prefixed with this prefix.
125 suffix | yes | string | Each node of the cabinet will be suffixed with this suffix
126 radical | yes | string | Regexp used to generate cabinet nodes name. Syntax: "10-20" will give you 11 machines numbered from 10 to 20, "10-20;2" will give you 12 machines, one with the number 2, others numbered as before. The produced number is concatenated between prefix and suffix to form machine names.
127 speed | yes | int | Same as the ``speed`` attribute of the @ref pf_tag_host "<host>" tag.
128 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the @ref pf_tag_link "link section" for syntax/details.
129 lat | yes | int | Latency for the links between nodes and backbone (if any). See the @ref pf_tag_link "link section" for syntax/details.
132 Please note that as of now, it is impossible to change attributes such as,
133 amount of cores (always set to 1), the sharing policy of the links (always set to @ref pf_sharing_policy_splitduplex "SPLITDUPLEX").
137 The following example was taken from ``examples/platforms/meta_cluster.xml`` and
138 shows how to use the cabinet tag.
141 <zone id="my_cluster1" routing="Cluster">
142 <cabinet id="cabinet1" prefix="host-" suffix=".cluster1"
143 speed="1Gf" bw="125MBps" lat="100us" radical="1-10"/>
144 <cabinet id="cabinet2" prefix="host-" suffix=".cluster1"
145 speed="1Gf" bw="125MBps" lat="100us" radical="11-20"/>
146 <cabinet id="cabinet3" prefix="host-" suffix=".cluster1"
147 speed="1Gf" bw="125MBps" lat="100us" radical="21-30"/>
149 <backbone id="backbone1" bandwidth="2.25GBps" latency="500us"/>
154 Please note that you must specify the @ref pf_backbone "<backbone>"
155 tag by yourself; this is not done automatically and there are no checks
156 that ensure this backbone was defined.
158 The hosts generated in the above example are named host-1.cluster, host-2.cluster1
162 @subsection pf_ne Network equipments
164 There are two tags at all times available to represent network entities and
165 several other tags that are available only in certain contexts.
166 1. ``<link>``: Represents a entity that has a limited bandwidth, a
167 latency, and that can be shared according to TCP way to share this
170 The concept of links in SimGrid may not be intuitive, as links are not
171 limited to connecting (exactly) two entities; in fact, you can have more than
172 two equipments connected to it. (In graph theoretical terms: A link in
173 SimGrid is not an edge, but a hyperedge)
175 2. ``<router/>``: Represents an entity that a message can be routed
176 to, but that is unable to execute any code. In SimGrid, routers have also
177 no impact on the performance: Routers do not limit any bandwidth nor
178 do they increase latency. As a matter of fact, routers are (almost) ignored
179 by the simulator when the simulation has begun.
181 3. ``<backbone/>``: This tag is only available when the containing network zone is
182 used as a cluster (i.e., mode="Cluster")
185 If you want to represent an entity like a switch, you must use ``<link>`` (see section). Routers are used
186 to run some routing algorithm and determine routes (see Section @ref pf_routing for details).
188 @subsubsection pf_backbone <backbone/>
191 This tag is <b>only available</b> when the containing network zone uses the "Cluster" routing mode!
193 Using this tag, you can designate an already existing link to be a backbone.
195 Attribute name | Mandatory | Values | Description
196 --------------- | --------- | ------ | -----------
197 id | yes | string | Name of the link that is supposed to act as a backbone.
199 @subsection pf_storage Storage
202 This is a prototype version that should evolve quickly, hence this
203 is just some doc valuable only at the time of writing.
204 This section describes the storage management under SimGrid ; nowadays
205 it's only usable with MSG. It relies basically on linux-like concepts.
206 You also may want to have a look to its corresponding section in
207 @ref msg_file ; access functions are organized as a POSIX-like
210 @subsubsection pf_sto_conc Storage - Main Concepts
212 The storage facilities implemented in SimGrid help to model (and account for)
213 storage devices, such as tapes, hard-drives, CD or DVD devices etc.
214 A typical situation is depicted in the figure below:
216 @image html ./webcruft/storage_sample_scenario.png
217 @image latex ./webcruft/storage_sample_scenario.png "storage_sample_scenario" width=@textwidth
219 In this figure, two hosts called Bob and Alice are interconnected via a network
220 and each host is physically attached to a disk; it is not only possible for each host to
221 mount the disk they are attached to directly, but they can also mount disks
222 that are in a remote location. In this example, Bob mounts Alice's disk remotely
223 and accesses the storage via the network.
225 SimGrid provides 3 different entities that can be used to model setups
226 that include storage facilities:
228 Entity name | Description
229 --------------- | -----------
230 @ref pf_storage_entity_storage_type "storage_type" | Defines a template for a particular kind of storage (such as a hard-drive) and specifies important features of the storage, such as capacity, performance (read/write), contents, ... Different models of hard-drives use different storage_types (because the difference between an SSD and an HDD does matter), as they differ in some specifications (e.g., different sizes or read/write performance).
231 @ref pf_tag_storage "storage" | Defines an actual instance of a storage type (disk, RAM, ...); uses a ``storage_type`` template (see line above) so that you don't need to re-specify the same details over and over again.
232 @ref pf_tag_mount "mount" | Must be wrapped by a @ref pf_tag_host tag; declares which storage(s) this host has mounted and where (i.e., the mountpoint).
235 @anchor pf_storage_content_file
236 ### %Storage Content File ###
238 In order to assess exactly how much time is spent reading from the storage,
239 SimGrid needs to know what is stored on the storage device (identified by distinct (file-)name, like in a file system)
240 and what size this content has.
243 The content file is never changed by the simulation; it is parsed once
244 per simulation and kept in memory afterwards. When the content of the
245 storage changes, only the internal SimGrid data structures change.
247 @anchor pf_storage_content_file_structure
248 #### Structure of a %Storage Content File ####
250 Here is an excerpt from two storage content file; if you want to see the whole file, check
251 the file ``examples/platforms/content/storage_content.txt`` that comes with the
254 SimGrid essentially supports two different formats: UNIX-style filepaths should
255 follow the well known format:
258 /lib/libsimgrid.so.3.6.2 12710497
262 /bin/simgrid_update_xml 5018
263 /bin/graphicator 66986
264 /bin/simgrid-colorizer 2993
269 Windows filepaths, unsurprisingly, use the windows style:
272 @Windows@avastSS.scr 41664
273 @Windows@bfsvc.exe 75264
274 @Windows@bootstat.dat 67584
275 @Windows@CoreSingleLanguage.xml 31497
277 @Windows@dchcfg64.exe 335464
278 @Windows@dcmdev64.exe 93288
282 The different file formats come at a cost; in version 3.12 (and most likely
283 in later versions, too), copying files from windows-style storages to unix-style
284 storages (and vice versa) is not supported.
286 @anchor pf_storage_content_file_create
287 #### Generate a %Storage Content File ####
289 If you want to generate a storage content file based on your own filesystem (or at least a filesystem you have access to),
290 try running this command (works only on unix systems):
293 find . -type f -exec ls -1s --block=1 {} @; 2>/dev/null | awk '{ print $2 " " $1}' > ./content.txt
296 @subsubsection pf_storage_entities The Storage Entities
298 These are the entities that you can use in your platform files to include
299 storage in your model. See also the list of our @ref pf_storage_example_files "example files";
300 these might also help you to get started.
302 @anchor pf_storage_entity_storage_type
303 #### @<storage_type@> ####
305 Attribute name | Mandatory | Values | Description
306 --------------- | --------- | ------ | -----------
307 id | yes | string | Identifier of this storage_type; used when referring to it
308 model | no | string | In the future, this will allow to change the performance model to use
309 size | yes | string | Specifies the amount of available storage space; you can specify storage like "500GiB" or "500GB" if you want. (TODO add a link to all the available abbreviations)
310 content | yes | string | Path to a @ref pf_storage_content_file "Storage Content File" on your system. This file must exist.
312 This tag must contain some predefined model properties, specified via the <model_prop> tag. Here is a list,
313 see below for an example:
315 Property id | Mandatory | Values | Description
316 --------------- | --------- | ------ | -----------
317 Bwrite | yes | string | Bandwidth for write access; in B/s (but you can also specify e.g. "30MBps")
318 Bread | yes | string | Bandwidth for read access; in B/s (but you can also specify e.g. "30MBps")
321 A storage_type can also contain the <b><prop></b> tag. The <prop> tag allows you
322 to associate additional information to this <storage_type> and follows the
323 attribute/value schema; see the example below. You may want to use it to give information to
324 the tool you use for rendering your simulation, for example.
326 Here is a complete example for the ``storage_type`` tag:
328 <storage_type id="single_HDD" size="4000">
329 <model_prop id="Bwrite" value="30MBps" />
330 <model_prop id="Bread" value="100MBps" />
331 <prop id="Brand" value="Western Digital" />
335 @subsubsection pf_tag_storage <storage>
337 Attributes | Mandatory | Values | Description
338 -------------- | --------- | ------ | -----------
339 id | yes | string | Identifier of this ``storage``; used when referring to it
340 typeId | yes | string | Here you need to refer to an already existing @ref pf_storage_entity_storage_type "@<storage_type@>"; the storage entity defined by this tag will then inherit the properties defined there.
341 attach | yes | string | Name of a host (see Section @ref pf_tag_host) to which this storage is <i>physically</i> attached to (e.g., a hard drive in a computer)
342 content | no | string | When specified, overwrites the content attribute of @ref pf_storage_entity_storage_type "@<storage_type@>"
344 Here are two examples:
347 <storage id="Disk1" typeId="single_HDD" attach="bob" />
349 <storage id="Disk2" typeId="single_SSD"
350 content="content/win_storage_content.txt" />
353 The first example is straightforward: A disk is defined and called "Disk1"; it is
354 of type "single_HDD" (shown as an example of @ref pf_storage_entity_storage_type "@<storage_type@>" above) and attached
355 to a host called "bob" (the definition of this host is omitted here).
357 The second storage is called "Disk2", is still of the same type as Disk1 but
358 now specifies a new content file (so the contents will be different from Disk1)
359 and the filesystem uses the windows style; finally, it is attached to a second host,
360 called alice (which is again not defined here).
362 @subsubsection pf_tag_mount <mount>
364 | Attribute | Mandatory | Values | Description |
365 | ----------- | ----------- | -------- | ------------- |
366 | id | yes | string | Refers to a @ref pf_tag_storage "<storage>" entity that will be mounted on that computer |
367 | name | yes | string | Path/location to/of the logical reference (mount point) of this disk
369 This tag must be enclosed by a @ref pf_tag_host tag. It then specifies where the mountpoint of a given storage device (defined by the ``id`` attribute)
370 is; this location is specified by the ``name`` attribute.
372 Here is a simple example, taken from the file ``examples/platform/storage.xml``:
375 <storage_type id="single_SSD" size="500GiB">
376 <model_prop id="Bwrite" value="60MBps" />
377 <model_prop id="Bread" value="200MBps" />
380 <storage id="Disk2" typeId="single_SSD"
381 content="content/win_storage_content.txt"
383 <storage id="Disk4" typeId="single_SSD"
384 content="content/small_content.txt"
387 <host id="alice" speed="1Gf">
388 <mount storageId="Disk2" name="c:"/>
391 <host id="denise" speed="1Gf">
392 <mount storageId="Disk2" name="c:"/>
393 <mount storageId="Disk4" name="/home"/>
397 This example is quite interesting, as the same device, called "Disk2", is mounted by
398 two hosts at the same time! Note, however, that the host called ``alice`` is actually
399 attached to this storage, as can be seen in the @ref pf_tag_storage "<storage>"
400 tag. This means that ``denise`` must access this storage through the network, but SimGrid automatically takes
401 care of that for you.
403 Furthermore, this example shows that ``denise`` has mounted two storages with different
404 filesystem types (unix and windows). In general, a host can mount as many storage devices as
408 Again, the difference between ``attach`` and ``mount`` is simply that
409 an attached storage is always physically inside (or connected to) that machine;
410 for instance, a USB stick is attached to one and only one machine (where it's plugged-in)
411 but it can only be mounted on others, as mounted storage can also be a remote location.
413 ###### Example files #####
415 @verbinclude example_filelist_xmltag_mount
417 @subsubsection pf_storage_example_files Example files
419 Several examples were already discussed above; if you're interested in full examples,
420 check the the following platforms:
422 1. ``examples/platforms/storage.xml``
423 2. ``examples/platforms/remote_io.xml``
425 If you're looking for some examplary C code, you may find the source code
426 available in the directory ``examples/msg/io/`` useful.
428 @subsubsection pf_storage_examples_modelling Modelling different situations
430 The storage functionality of SimGrid is type-agnostic, that is, the implementation
431 does not presume any type of storage, such as HDDs/SSDs, RAM,
432 CD/DVD devices, USB sticks etc.
434 This allows the user to apply the simulator for a wide variety of scenarios; one
435 common scenario would be the access of remote RAM.
437 #### Modelling the access of remote RAM ####
439 How can this be achieved in SimGrid? Let's assume we have a setup where three hosts
440 (HostA, HostB, HostC) need to access remote RAM:
450 An easy way to model this scenario is to setup and define the RAM via the
451 @ref pf_tag_storage "storage" and @ref pf_storage_entity_storage_type "storage type"
452 entities and attach it to a remote dummy host; then, every host can have their own links
453 to this host (modelling for instance certain scenarios, such as PCIe ...)
458 RAM - Dummy -- Host B
463 Now, if read from this storage, the host that mounts this storage
464 communicates to the dummy host which reads from RAM and
465 sends the information back.
468 @section pf_routing Routing
470 To achieve high performance, the routing tables used within SimGrid are
471 static. This means that routing between two nodes is calculated once
472 and will not change during execution. The SimGrid team chose to use this
473 approach as it is rare to have a real deficiency of a resource;
474 most of the time, a communication fails because the links experience too much
475 congestion and hence, your connection stops before the timeout or
476 because the computer designated to be the destination of that message
479 We also chose to use shortest paths algorithms in order to emulate
480 routing. Doing so is consistent with the reality: [RIP](https://en.wikipedia.org/wiki/Routing_Information_Protocol),
481 [OSPF](https://en.wikipedia.org/wiki/Open_Shortest_Path_First), [BGP](https://en.wikipedia.org/wiki/Border_Gateway_Protocol)
482 are all calculating shortest paths. They do require some time to converge, but
483 eventually, when the routing tables have stabilized, your packets will follow
486 @subsection pf_tag_zone <zone>
488 @subsection pf_rm Routing models
490 For each network zone, you must define explicitly which routing model will
491 be used. There are 3 different categories for routing models:
493 1. @ref pf_routing_model_shortest_path "Shortest-path" based models: SimGrid calculates shortest
494 paths and manages them. Behaves more or less like most real life
496 2. @ref pf_routing_model_manual "Manually-entered" route models: you have to define all routes
497 manually in the platform description file; this can become
498 tedious very quickly, as it is very verbose.
499 Consistent with some manually managed real life routing.
500 3. @ref pf_routing_model_simple "Simple/fast models": those models offer fast, low memory routing
501 algorithms. You should consider to use this type of model if
502 you can make some assumptions about your network zone.
503 Routing in this case is more or less ignored.
505 @subsubsection pf_raf The router affair
507 Using routers becomes mandatory when using shortest-path based
508 models or when using the bindings to the ns-3 packet-level
509 simulator instead of the native analytical network model implemented
512 For graph-based shortest path algorithms, routers are mandatory, because these
513 algorithms require a graph as input and so we need to have source and
514 destination for each edge.
516 Routers are naturally an important concept ns-3 since the
517 way routers run the packet routing algorithms is actually simulated.
518 SimGrid's analytical models however simply aggregate the routing time
519 with the transfer time.
521 So why did we incorporate routers in SimGrid? Rebuilding a graph representation
522 only from the route information turns out to be a very difficult task, because
523 of the missing information about how routes intersect. That is why we
524 introduced routers, which are simply used to express these intersection points.
525 It is important to understand that routers are only used to provide topological
528 To express this topological information, a <b>route</b> has to be
529 defined in order to declare which link is connected to a router.
532 @subsubsection pf_routing_model_shortest_path Shortest-path based models
534 The following table shows all the models that compute routes using
535 shortest-paths algorithms are currently available in SimGrid. More detail on how
536 to choose the best routing model is given in the Section called @"@ref pf_routing_howto_choose_wisely@".
538 | Name | Description |
539 | --------------------------------------------------- | -------------------------------------------------------------------------- |
540 | @ref pf_routing_model_floyd "Floyd" | Floyd routing data. Pre-calculates all routes once |
541 | @ref pf_routing_model_dijkstra "Dijkstra" | Dijkstra routing data. Calculates routes only when needed |
542 | @ref pf_routing_model_dijkstracache "DijkstraCache" | Dijkstra routing data. Handles some cache for already calculated routes. |
544 All those shortest-path models are instanciated in the same way and are
545 completely interchangeable. Here are some examples:
547 @anchor pf_routing_model_floyd
552 <zone id="zone0" routing="Floyd">
554 <cluster id="my_cluster_1" prefix="c-" suffix=""
555 radical="0-1" speed="1000000000" bw="125000000" lat="5E-5"
556 router_id="router1"/>
558 <zone id="zone1" routing="None">
559 <host id="host1" speed="1000000000"/>
562 <link id="link1" bandwidth="100000" latency="0.01"/>
564 <zoneroute src="my_cluster_1" dst="zone1"
567 <link_ctn id="link1"/>
573 zoneroute given at the end gives a topological information: link1 is
574 between router1 and host1.
576 #### Example platform files ####
578 This is an automatically generated list of example files that use the Floyd
579 routing model (the path is given relative to SimGrid's source directory)
581 @verbinclude example_filelist_routing_floyd
583 @anchor pf_routing_model_dijkstra
586 #### Example platform files ####
588 This is an automatically generated list of example files that use the Dijkstra
589 routing model (the path is given relative to SimGrid's source directory)
591 @verbinclude example_filelist_routing_dijkstra
595 <zone id="zone_2" routing="Dijkstra">
596 <host id="zone_2_host1" speed="1000000000"/>
597 <host id="zone_2_host2" speed="1000000000"/>
598 <host id="zone_2_host3" speed="1000000000"/>
599 <link id="zone_2_link1" bandwidth="1250000000" latency="5E-4"/>
600 <link id="zone_2_link2" bandwidth="1250000000" latency="5E-4"/>
601 <link id="zone_2_link3" bandwidth="1250000000" latency="5E-4"/>
602 <link id="zone_2_link4" bandwidth="1250000000" latency="5E-4"/>
603 <router id="central_router"/>
604 <router id="zone_2_gateway"/>
605 <!-- routes providing topological information -->
606 <route src="central_router" dst="zone_2_host1"><link_ctn id="zone_2_link1"/></route>
607 <route src="central_router" dst="zone_2_host2"><link_ctn id="zone_2_link2"/></route>
608 <route src="central_router" dst="zone_2_host3"><link_ctn id="zone_2_link3"/></route>
609 <route src="central_router" dst="zone_2_gateway"><link_ctn id="zone_2_link4"/></route>
613 @anchor pf_routing_model_dijkstracache
614 ### DijkstraCache ###
616 DijkstraCache example:
618 <zone id="zone_2" routing="DijkstraCache">
619 <host id="zone_2_host1" speed="1000000000"/>
621 (platform unchanged compared to upper example)
624 #### Example platform files ####
626 This is an automatically generated list of example files that use the DijkstraCache
627 routing model (the path is given relative to SimGrid's source directory):
629 Editor's note: At the time of writing, no platform file used this routing model - so
630 if there are no example files listed here, this is likely to be correct.
632 @verbinclude example_filelist_routing_dijkstra_cache
634 @subsubsection pf_routing_model_manual Manually-entered route models
636 | Name | Description |
637 | ---------------------------------- | ------------------------------------------------------------------------------ |
638 | @ref pf_routing_model_full "Full" | You have to enter all necessary routers manually; that is, every single route. This may consume a lot of memory when the XML is parsed and might be tedious to write; i.e., this is only recommended (if at all) for small platforms. |
640 @anchor pf_routing_model_full
645 <zone id="zone0" routing="Full">
646 <host id="host1" speed="1000000000"/>
647 <host id="host2" speed="1000000000"/>
648 <link id="link1" bandwidth="125000000" latency="0.000100"/>
649 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
653 #### Example platform files ####
655 This is an automatically generated list of example files that use the Full
656 routing model (the path is given relative to SimGrid's source directory):
658 @verbinclude example_filelist_routing_full
660 @subsubsection pf_routing_model_simple Simple/fast models
662 | Name | Description |
663 | ---------------------------------------- | ------------------------------------------------------------------------------ |
664 | @ref pf_routing_model_cluster "Cluster" | This is specific to the @ref pf_tag_cluster "<cluster/>" tag and should not be used by the user, as several assumptions are made. |
665 | @ref pf_routing_model_none "None" | No routing at all. Unless you know what you're doing, avoid using this mode in combination with a non-constant network model. |
666 | @ref pf_routing_model_vivaldi "Vivaldi" | Perfect when you want to use coordinates. Also see the corresponding @ref pf_P2P_tags "P2P section" below. |
668 @anchor pf_routing_model_cluster
672 In this mode, the @ref pf_cabinet "<cabinet/>" tag is available.
674 #### Example platform files ####
676 This is an automatically generated list of example files that use the Cluster
677 routing model (the path is given relative to SimGrid's source directory):
679 @verbinclude example_filelist_routing_cluster
681 @anchor pf_routing_model_none
685 This model does exactly what it's name advertises: Nothing. There is no routing
686 available within this model and if you try to communicate within the zone that
687 uses this model, SimGrid will fail unless you have explicitly activated the
688 @ref options_model_select_network_constant "Constant Network Model" (this model charges
689 the same for every single communication). It should
690 be noted, however, that you can still attach an @ref pf_tag_zoneroute "ZoneRoute",
691 as is demonstrated in the example below:
693 @verbinclude platforms/cluster_and_one_host.xml
695 #### Example platform files ####
697 This is an automatically generated list of example files that use the None
698 routing model (the path is given relative to SimGrid's source directory):
700 @verbinclude example_filelist_routing_none
703 @anchor pf_routing_model_vivaldi
706 For more information on how to use the [Vivaldi Coordinates](https://en.wikipedia.org/wiki/Vivaldi_coordinates),
707 see also Section @ref pf_P2P_tags "P2P tags".
709 Note that it is possible to combine the Vivaldi routing model with other routing models;
710 an example can be found in the file @c examples/platforms/cloud.xml. This
711 examples models a NetZone using Vivaldi that contains other NetZones that use different
714 #### Example platform files ####
716 This is an automatically generated list of example files that use the None
717 routing model (the path is given relative to SimGrid's source directory):
719 @verbinclude example_filelist_routing_vivaldi
722 @subsection ps_dec Defining routes
724 There are currently four different ways to define routes:
726 | Name | Description |
727 | ------------------------------------------------- | ----------------------------------------------------------------------------------- |
728 | @ref pf_tag_route "route" | Used to define route between host/router |
729 | @ref pf_tag_zoneroute "zoneRoute" | Used to define route between different zones |
730 | @ref pf_tag_bypassroute "bypassRoute" | Used to supersede normal routes as calculated by the network model between host/router; e.g., can be used to use a route that is not the shortest path for any of the shortest-path routing models. |
731 | @ref pf_tag_bypassasroute "bypassZoneRoute" | Used in the same way as bypassRoute, but for zones |
733 Basically all those tags will contain an (ordered) list of references
734 to link that compose the route you want to define.
736 Consider the example below:
739 <route src="Alice" dst="Bob">
740 <link_ctn id="link1"/>
741 <link_ctn id="link2"/>
742 <link_ctn id="link3"/>
746 The route here from host Alice to Bob will be first link1, then link2,
747 and finally link3. What about the reverse route? @ref pf_tag_route "Route" and
748 @ref pf_tag_zoneroute "zoneroute" have an optional attribute @c symmetrical, that can
749 be either @c YES or @c NO. @c YES means that the reverse route is the same
750 route in the inverse order, and is set to @c YES by default. Note that
751 this is not the case for bypass*Route, as it is more probable that you
752 want to bypass only one default route.
754 For an @ref pf_tag_zoneroute "zoneroute", things are just slightly more complicated, as you have
755 to give the id of the gateway which is inside the zone you want to access ...
756 So it looks like this:
759 <zoneroute src="zone1" dst="zone2"
760 gw_src="router1" gw_dst="router2">
761 <link_ctn id="link1"/>
765 gw == gateway, so when any message are trying to go from zone1 to zone2,
766 it means that it must pass through router1 to get out of the zone, then
767 pass through link1, and get into zone2 by being received by router2.
768 router1 must belong to zone1 and router2 must belong to zone2.
770 @subsubsection pf_tag_zoneroute <zoneRoute>
772 The purpose of this entity is to define a route between two
773 NetZones. Recall that all zones form a tree, so to connect two
774 sibiling zones, you must give such a zoneRoute specifying the source
775 and destination zones, along with the gateway in each zone (ie, the
776 point to reach within that zone to reach the netzone), and the list of
777 links in the ancestor zone to go from one zone to another.
779 So, to go from an host @c src_host that is within zone @c src, to an
780 host @c dst_host that is within @c dst, you need to:
782 - move within zone @c src, from @c src_host to the specified @c gw_src;
783 - traverse all links specified by the zoneRoute (they are supposed to be within the common ancestor);
784 - move within zone @c dst, from @c gw_dst to @c dst_host.
788 | Attribute name | Mandatory | Values | Description |
789 | --------------- | --------- | ------ | ----------- |
790 | src | yes | String | The identifier of the source zone |
791 | dst | yes | String | See the @c src attribute |
792 | gw_src | yes | String | The gateway that will be used within the src zone; this can be any @ref pf_tag_host "Host" or @ref pf_router "Router" defined within the src zone. |
793 | gw_dst | yes | String | Same as @c gw_src, but with the dst zone instead. |
794 | symmetrical | no | YES@|NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
799 <zone id="zone0" routing="Full">
800 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
801 radical="0-149" speed="1000000000" bw="125000000" lat="5E-5"
802 bb_bw="2250000000" bb_lat="5E-4"/>
804 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
805 radical="150-299" speed="1000000000" bw="125000000" lat="5E-5"
806 bb_bw="2250000000" bb_lat="5E-4"/>
808 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
810 <zoneroute src="my_cluster_1" dst="my_cluster_2"
811 gw_src="c-my_cluster_1_router.me"
812 gw_dst="c-my_cluster_2_router.me">
813 <link_ctn id="backbone"/>
815 <zoneroute src="my_cluster_2" dst="my_cluster_1"
816 gw_src="c-my_cluster_2_router.me"
817 gw_dst="c-my_cluster_1_router.me">
818 <link_ctn id="backbone"/>
823 @subsubsection pf_tag_route <route>
825 The principle is the same as for
826 @ref pf_tag_zoneroute "ZoneRoute": The route contains a list of links that
827 provide a path from @c src to @c dst. Here, @c src and @c dst can both be either a
828 @ref pf_tag_host "host" or @ref pf_router "router". This is mostly useful for the
829 @ref pf_routing_model_full "Full routing model" as well as for the
830 @ref pf_routing_model_shortest_path "shortest-paths" based models (as they require
831 topological information).
834 | Attribute name | Mandatory | Values | Description |
835 | --------------- | --------- | ---------------------- | ----------- |
836 | src | yes | String | The value given to the source's "id" attribute |
837 | dst | yes | String | The value given to the destination's "id" attribute. |
838 | symmetrical | no | YES@| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
843 A route in the @ref pf_routing_model_full "Full routing model" could look like this:
845 <route src="Tremblay" dst="Bourassa">
846 <link_ctn id="4"/><link_ctn id="3"/><link_ctn id="2"/><link_ctn id="0"/><link_ctn id="1"/><link_ctn id="6"/><link_ctn id="7"/>
850 A route in the @ref pf_routing_model_shortest_path "Shortest-Path routing model" could look like this:
852 <route src="Tremblay" dst="Bourassa">
857 You must only have one link in your routes when you're using them to provide
858 topological information, as the routes here are simply the edges of the
859 (network-)graph and the employed algorithms need to know which edge connects
860 which pair of entities.
862 @subsubsection pf_tag_bypassasroute bypasszoneroute
864 As said before, once you choose
865 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
866 define some of your routes, which will be specific. You may also want
867 to bypass some routes defined in lower level zone at an upper stage:
868 <b>bypasszoneroute</b> is the tag you're looking for. It allows to
869 bypass routes defined between already defined between zone (if you want
870 to bypass route for a specific host, you should just use byPassRoute).
871 The principle is the same as zoneroute: <b>bypasszoneroute</b> contains
872 list of links that are in the path between src and dst.
876 | Attribute name | Mandatory | Values | Description |
877 | --------------- | --------- | ---------------------- | ----------- |
878 | src | yes | String | The value given to the source zone's "id" attribute |
879 | dst | yes | String | The value given to the destination zone's "id" attribute. |
880 | gw_src | yes | String | The value given to the source gateway's "id" attribute; this can be any host or router within the src zone |
881 | gw_dst | yes | String | The value given to the destination gateway's "id" attribute; this can be any host or router within the dst zone|
882 | symmetrical | no | YES@| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
887 <bypasszoneRoute src="my_cluster_1" dst="my_cluster_2"
888 gw_src="my_cluster_1_router"
889 gw_dst="my_cluster_2_router">
890 <link_ctn id="link_tmp"/>
894 This example shows that link @c link_tmp (definition not displayed here) directly
895 connects the router @c my_cluster_1_router in the source cluster to the router
896 @c my_cluster_2_router in the destination router. Additionally, as the @c symmetrical
897 attribute was not given, this route is presumed to be symmetrical.
899 @subsubsection pf_tag_bypassroute bypassRoute
901 As said before, once you choose
902 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
903 define some of your routes, which will be specific. You may also want
904 to bypass some routes defined in lower level zone at an upper stage:
905 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
906 routes defined between <b>host/router</b>. The principle is the same
907 as route: <b>bypassRoute</b> contains list of links references of
908 links that are in the path between src and dst.
912 | Attribute name | Mandatory | Values | Description |
913 | --------------- | --------- | ---------------------- | ----------- |
914 | src | yes | String | The value given to the source zone's "id" attribute |
915 | dst | yes | String | The value given to the destination zone's "id" attribute. |
916 | symmetrical | no | YES @| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
921 <bypassRoute src="host_1" dst="host_2">
922 <link_ctn id="link_tmp"/>
926 This example shows that link @c link_tmp (definition not displayed here) directly
927 connects host @c host_1 to host @c host_2. Additionally, as the @c symmetrical
928 attribute was not given, this route is presumed to be symmetrical.
930 @subsection pb_baroex Basic Routing Example
932 Let's say you have an zone named zone_Big that contains two other zone, zone_1
933 and zone_2. If you want to make a host (h1) from zone_1 with another one
934 (h2) from zone_2 then you'll have to proceed as follows:
935 @li First, you have to ensure that a route is defined from h1 to the
936 zone_1's exit gateway and from h2 to zone_2's exit gateway.
937 @li Then, you'll have to define a route between zone_1 to zone_2. As those
938 zone are both resources belonging to zone_Big, then it has to be done
939 at zone_big level. To define such a route, you have to give the
940 source zone (zone_1), the destination zone (zone_2), and their respective
941 gateway (as the route is effectively defined between those two
942 entry/exit points). Elements of this route can only be elements
943 belonging to zone_Big, so links and routers in this route should be
944 defined inside zone_Big. If you choose some shortest-path model,
945 this route will be computed automatically.
947 As said before, there are mainly 2 tags for routing:
948 @li <b>zoneroute</b>: to define routes between two <b>zone</b>
949 @li <b>route</b>: to define routes between two <b>host/router</b>
951 As we are dealing with routes between zone, it means that those we'll
952 have some definition at zone_Big level. Let consider zone_1 contains 1
953 host, 1 link and one router and zone_2 3 hosts, 4 links and one router.
954 There will be a central router, and a cross-like topology. At the end
955 of the crosses arms, you'll find the 3 hosts and the router that will
956 act as a gateway. We have to define routes inside those two zone. Let
957 say that zone_1 contains full routes, and zone_2 contains some Floyd
958 routing (as we don't want to bother with defining all routes). As
959 we're using some shortest path algorithms to route into zone_2, we'll
960 then have to define some <b>route</b> to gives some topological
961 information to SimGrid. Here is a file doing it all:
964 <zone id="zone_Big" routing="Dijkstra">
965 <zone id="zone_1" routing="Full">
966 <host id="zone_1_host1" speed="1000000000"/>
967 <link id="zone_1_link" bandwidth="1250000000" latency="5E-4"/>
968 <router id="zone_1_gateway"/>
969 <route src="zone_1_host1" dst="zone_1_gateway">
970 <link_ctn id="zone_1_link"/>
973 <zone id="zone_2" routing="Floyd">
974 <host id="zone_2_host1" speed="1000000000"/>
975 <host id="zone_2_host2" speed="1000000000"/>
976 <host id="zone_2_host3" speed="1000000000"/>
977 <link id="zone_2_link1" bandwidth="1250000000" latency="5E-4"/>
978 <link id="zone_2_link2" bandwidth="1250000000" latency="5E-4"/>
979 <link id="zone_2_link3" bandwidth="1250000000" latency="5E-4"/>
980 <link id="zone_2_link4" bandwidth="1250000000" latency="5E-4"/>
981 <router id="central_router"/>
982 <router id="zone_2_gateway"/>
983 <!-- routes providing topological information -->
984 <route src="central_router" dst="zone_2_host1"><link_ctn id="zone_2_link1"/></route>
985 <route src="central_router" dst="zone_2_host2"><link_ctn id="zone_2_link2"/></route>
986 <route src="central_router" dst="zone_2_host3"><link_ctn id="zone_2_link3"/></route>
987 <route src="central_router" dst="zone_2_gateway"><link_ctn id="zone_2_link4"/></route>
989 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
991 <zoneroute src="zone_1" dst="zone_2"
992 gw_src="zone_1_gateway"
993 gw_dst="zone_2_gateway">
994 <link_ctn id="backbone"/>
999 @section pf_other Other tags
1001 The following tags can be used inside a @<platform@> tag even if they are not
1002 directly describing the platform:
1004 - @ref pf_tag_config passes configuration options, e.g. to change the network model;
1005 - @ref pf_tag_prop gives user-defined properties to various elements
1007 @subsection pf_trace trace and trace_connect
1009 Both tags are an alternate way to pass files containing information on
1010 availability, state etc. to an entity. (See also @ref howto_churn).
1011 Instead of referring to the file directly in the host, link, or
1012 cluster tag, you proceed by defining a trace with an id corresponding
1013 to a file, later a host/link/cluster, and finally using trace_connect
1014 you say that the file trace must be used by the entity.
1020 <zone id="zone0" routing="Full">
1021 <host id="bob" speed="1000000000"/>
1023 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1024 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1028 The order here is important. @c trace_connect must come
1029 after the elements @c trace and @c host, as both the host
1030 and the trace definition must be known when @c trace_connect
1031 is parsed; the order of @c trace and @c host is arbitrary.
1034 #### @c trace attributes ####
1037 | Attribute name | Mandatory | Values | Description |
1038 | --------------- | --------- | ---------------------- | ----------- |
1039 | id | yes | String | Identifier of this trace; this is the name you pass on to @c trace_connect. |
1040 | file | no | String | Filename of the file that contains the information - the path must follow the style of your OS. You can omit this, but then you must specifiy the values inside of <trace> and </trace> - see the example below. |
1041 | trace_periodicity | yes | String | This is the same as for @ref pf_tag_host "hosts" (see there for details) |
1043 Here is an example of trace when no file name is provided:
1046 <trace id="myTrace" periodicity="1.0">
1053 #### @c trace_connect attributes ####
1055 | Attribute name | Mandatory | Values | Description |
1056 | --------------- | --------- | ---------------------- | ----------- |
1057 | kind | no | HOST_AVAIL@|POWER@|<br/>LINK_AVAIL@|BANDWIDTH@|LATENCY (Default: HOST_AVAIL) | Describes the kind of trace. |
1058 | trace | yes | String | Identifier of the referenced trace (specified of the trace's @c id attribute) |
1059 | element | yes | String | The identifier of the referenced entity as given by its @c id attribute |
1061 @section pf_hints Hints, tips and frequently requested features
1063 Now you should know at least the syntax and be able to create a
1064 platform by your own. However, after having ourselves wrote some platforms, there
1065 are some best practices you should pay attention to in order to
1066 produce good platform and some choices you can make in order to have
1067 faster simulations. Here's some hints and tips, then.
1069 @subsection pf_hints_search Finding the platform example that you need
1071 Most platform files that we ship are in the @c examples/platforms
1072 folder. The good old @c grep tool can find the examples you need when
1073 wondering on a specific XML tag. Here is an example session searching
1074 for @ref pf_trace "trace_connect":
1077 % cd examples/platforms
1078 % grep -R -i -n --include="*.xml" "trace_connect" .
1079 ./two_hosts_platform_with_availability_included.xml:26:<trace_connect kind="SPEED" trace="A" element="Cpu A"/>
1080 ./two_hosts_platform_with_availability_included.xml:27:<trace_connect kind="HOST_AVAIL" trace="A_failure" element="Cpu A"/>
1081 ./two_hosts_platform_with_availability_included.xml:28:<trace_connect kind="SPEED" trace="B" element="Cpu B"/>
1082 ./two_hosts.xml:17: <trace_connect trace="Tremblay_power" element="Tremblay" kind="SPEED"/>
1085 @subsection pf_hint_generating How to generate different platform files?
1087 This is actually a good idea to search for a better platform file,
1088 that better fit the need of your study. To be honest, the provided
1089 examples are not representative of anything. They exemplify our XML
1090 syntax, but that's all. small_platform.xml for example was generated
1091 without much thought beyond that.
1093 The best thing to do when possible is to write your own platform file,
1094 that model the platform on which you run your code. For that, you
1095 could use <a href="https://gitlab.inria.fr/simgrid/platform-calibration">our
1096 calibration scripts</a>. This leads to very good fits between the
1097 platform, the model and the needs. The g5k.xml example resulted of
1098 such an effort, which also lead to <a href="https://github.com/lpouillo/topo5k/">an
1099 ongoing attempt</a> to automatically extract the SimGrid platform from
1100 the <a href="http://grid5000.fr/">Grid'5000</a> experimental platform.
1101 But it's hard to come up with generic models. Don't take these files
1102 too seriously. Actually, you should always challenge our models and
1103 your instanciation if the accuracy really matters to you (see <a
1104 href="https://hal.inria.fr/hal-00907887">this discussion</a>).
1106 But such advices only hold if you have a real platform and a real
1107 application at hand. It's moot for more abstract studies working on
1108 ideas and algorithms instead of technical artefacts. Well, in this
1109 case, there unfortunately is nothing better than this old and rusty
1110 <a href="http://pda.gforge.inria.fr/tools/download.html">simulacrum</a>.
1111 This project is dormant since over 10 years (and you will have to
1112 update the generated platforms with <tt>bin/simgrid_update_xml</tt> to
1113 use them), but that's the best we have for this right now....
1115 @subsection pf_zone_h Zone Hierarchy
1116 The network zone design allows SimGrid to go fast, because computing route is
1117 done only for the set of resources defined in the current zone. If you're using
1118 only a big zone containing all resource with no zone into it and you're
1119 using Full model, then ... you'll loose all interest into it. On the
1120 other hand, designing a binary tree of zone with, at the lower level,
1121 only one host, then you'll also loose all the good zone hierarchy can
1122 give you. Remind you should always be "reasonable" in your platform
1123 definition when choosing the hierarchy. A good choice if you try to
1124 describe a real life platform is to follow the zone described in
1125 reality, since this kind of trade-off works well for real life
1128 @subsection pf_exit_zone Exit Zone: why and how
1129 Users that have looked at some of our platforms may have notice a
1130 non-intuitive schema ... Something like that:
1134 <zone id="zone_4" routing="Full">
1135 <zone id="exitzone_4" routing="Full">
1136 <router id="router_4"/>
1138 <cluster id="cl_4_1" prefix="c_4_1-" suffix="" radical="1-20" speed="1000000000" bw="125000000" lat="5E-5" bb_bw="2250000000" bb_lat="5E-4"/>
1139 <cluster id="cl_4_2" prefix="c_4_2-" suffix="" radical="1-20" speed="1000000000" bw="125000000" lat="5E-5" bb_bw="2250000000" bb_lat="5E-4"/>
1140 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1141 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1142 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1143 <zoneroute src="cl_4_1"
1145 gw_src="c_4_1-cl_4_1_router"
1146 gw_dst="c_4_2-cl_4_2_router">
1147 <link_ctn id="4_1"/>
1148 <link_ctn id="bb_4"/>
1149 <link_ctn id="4_2"/>
1151 <zoneroute src="cl_4_1"
1153 gw_src="c_4_1-cl_4_1_router"
1155 <link_ctn id="4_1"/>
1156 <link_ctn id="bb_4"/>
1158 <zoneroute src="cl_4_2"
1160 gw_src="c_4_2-cl_4_2_router"
1162 <link_ctn id="4_2"/>
1163 <link_ctn id="bb_4"/>
1168 In the zone_4, you have an exitzone_4 defined, containing only one router,
1169 and routes defined to that zone from all other zone (as cluster is only a
1170 shortcut for an zone, see cluster description for details). If there was
1171 an upper zone, it would define routes to and from zone_4 with the gateway
1172 router_4. It's just because, as we did not allowed (for performances
1173 issues) to have routes from an zone to a single host/router, you have to
1174 enclose your gateway, when you have zone included in your zone, within an
1175 zone to define routes to it.
1177 @subsection pf_P2P_tags P2P or how to use coordinates
1178 SimGrid allows you to use some coordinated-based system, like vivaldi,
1179 to describe a platform. The main concept is that you have some peers
1180 that are located somewhere: this is the function of the
1181 <b>coordinates</b> of the @<peer@> or @<host@> tag. There's nothing
1182 complicated in using it, here is an example:
1185 <?xml version='1.0'?>
1186 <!DOCTYPE platform SYSTEM "https://simgrid.org/simgrid.dtd">
1187 <platform version="4">
1189 <zone id="zone0" routing="Vivaldi">
1190 <host id="100030591" coordinates="25.5 9.4 1.4" speed="1.5Gf" />
1191 <host id="100036570" coordinates="-12.7 -9.9 2.1" speed="7.3Gf" />
1193 <host id="100429957" coordinates="17.5 6.7 18.8" speed="8.3Gf" />
1198 Coordinates are then used to calculate latency (in microseconds)
1199 between two hosts by calculating the distance between the two hosts
1200 coordinates with the following formula: distance( (x1, y1, z1), (x2,
1201 y2, z2) ) = euclidian( (x1,y1), (x2,y2) ) + abs(z1) + abs(z2)
1203 In other words, we take the euclidian distance on the two first
1204 dimensions, and then add the absolute values found on the third
1205 dimension. This may seem strange, but it was found to allow better
1206 approximations of the latency matrices (see the paper describing
1209 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between zone.
1210 That is for example what is commonly done when using peers (see Section @ref pf_peer).
1212 <?xml version='1.0'?>
1213 <!DOCTYPE platform SYSTEM "https://simgrid.org/simgrid.dtd">
1214 <platform version="4">
1216 <zone id="zone0" routing="Vivaldi">
1217 <peer id="peer-0" coordinates="173.0 96.8 0.1" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1218 <peer id="peer-1" coordinates="247.0 57.3 0.6" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1219 <peer id="peer-2" coordinates="243.4 58.8 1.4" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1223 In such a case though, we connect the zone created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1224 This means that to route between zone1 and zone2, it will use the coordinates of router_zone1 and router_zone2.
1225 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1226 You may have noted that conveniently, a peer named FOO defines an zone named FOO and a router named router_FOO, which is why it works seamlessly with the <b>peer</b> tag.
1229 @subsection pf_routing_howto_choose_wisely Choosing wisely the routing model to use
1232 Choosing wisely the routing model to use can significantly fasten your
1233 simulation/save your time when writing the platform/save tremendous
1234 disk space. Here is the list of available model and their
1235 characteristics (lookup: time to resolve a route):
1237 @li <b>Full</b>: Full routing data (fast, large memory requirements,
1239 @li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1240 lookup, lesser memory requirements, shortest path routing only).
1241 Calculates all routes at once at the beginning.
1242 @li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1243 lookup, small memory requirements, shortest path routing only).
1244 Calculates a route when necessary.
1245 @li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1246 fast lookup, small memory requirements, shortest path routing
1247 only). Same as Dijkstra, except it handles a cache for latest used
1249 @li <b>None</b>: No routing (usable with Constant network only).
1250 Defines that there is no routes, so if you try to determine a
1251 route without constant network within this zone, SimGrid will raise
1253 @li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1254 @li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1257 @subsection pf_loopback I want to specify the characteristics of the loopback link!
1259 Each routing model automatically adds a loopback link for each declared host, i.e.,
1260 a network route from the host to itself, if no such route is declared in the XML
1261 file. This default link has a bandwidth of 498 Mb/s, a latency of 15 microseconds,
1262 and is <b>not</b> shared among network flows.
1264 If you want to specify the characteristics of the loopback link for a given host, you
1265 just have to specify a route from this host to itself with the desired characteristics
1266 in the XML file. This will prevent the routing model to add and use the default
1269 @subsection pf_switch I want to describe a switch but there is no switch tag!
1271 Actually we did not include switch tag. But when you're trying to
1272 simulate a switch, assuming
1273 fluid bandwidth models are used (which SimGrid uses by default unless
1274 ns-3 or constant network models are activated), the limiting factor is
1275 switch backplane bandwidth. So, essentially, at least from
1276 the simulation perspective, a switch is similar to a
1277 link: some device that is traversed by flows and with some latency and
1278 so,e maximum bandwidth. Thus, you can simply simulate a switch as a
1280 can be connected to this "switch", which is then included in routes just
1284 @subsection pf_multicabinets I want to describe multi-cabinets clusters!
1286 You have several possibilities, as usual when modeling things. If your
1287 cabinets are homogeneous and the intercabinet network negligible for
1288 your study, you should just create a larger cluster with all hosts at
1291 In the rare case where your hosts are not homogeneous between the
1292 cabinets, you can create your cluster completely manually. For that,
1293 create an As using the Cluster routing, and then use one
1294 <cabinet> for each cabinet. This cabinet tag can only be used an
1295 As using the Cluster routing schema, and creating
1297 Be warned that creating a cluster manually from the XML with
1298 <cabinet>, <backbone> and friends is rather tedious. The
1299 easiest way to retrieve some control of your model without diving into
1300 the <cluster> internals is certainly to create one separate
1301 <cluster> per cabinet and interconnect them together. This is
1302 what we did in the G5K example platform for the Graphen cluster.
1304 @subsection pf_platform_multipath I want to express multipath routing in platform files!
1306 It is unfortunately impossible to express the fact that there is more
1307 than one routing path between two given hosts. Let's consider the
1308 following platform file:
1311 <route src="A" dst="B">
1314 <route src="B" dst="C">
1317 <route src="A" dst="C">
1322 Although it is perfectly valid, it does not mean that data traveling
1323 from A to C can either go directly (using link 3) or through B (using
1324 links 1 and 2). It simply means that the routing on the graph is not
1325 trivial, and that data do not following the shortest path in number of
1326 hops on this graph. Another way to say it is that there is no implicit
1327 in these routing descriptions. The system will only use the routes you
1328 declare (such as <route src="A" dst="C"><link_ctn
1329 id="3"/></route>), without trying to build new routes by aggregating
1332 You are also free to declare platform where the routing is not
1333 symmetrical. For example, add the following to the previous file:
1336 <route src="C" dst="A">
1342 This makes sure that data from C to A go through B where data from A
1343 to C go directly. Don't worry about realism of such settings since
1344 we've seen ways more weird situation in real settings (in fact, that's
1345 the realism of very regular platforms which is questionable, but
1346 that's another story).