1 /*! \page platform Describing the virtual platform
5 In order to run any simulation, SimGrid must be provided with three things:
6 something to run (i.e., your code), a description of the platform on which you want to simulate your application, and
7 information about the deployment of the application: Which process should be executed onto which processor/core?
9 For the last two items, there are essentially three possible ways you can provide
10 this information as an input:
11 \li You can program, if you're using MSG, some of the platform and
12 deployment functions. If you choose to follow this approach, check the dedicated documentation
13 (\ref msg_simulation).
14 \li You can use two XML files: one for the platform description and the other for the deployment.
15 \li You can program the description of your platform in Lua format.
17 For more information on SimGrid's deployment features, please refer to the \ref deployment section.
19 The platform description may be intricate. This documentation is all about how to write this file. First, the basic
20 concepts are introduced. Then, advanced options are explained. Finally, some hints and tips on how to write a better
21 platform description are given.
23 \section pf_overview Some words about XML and DTD
25 We opted for XML not only because it is extensible but also because many tools (and plugins for existing tools) are
26 available that facilitate editing and validating XML files. Furthermore, libraries that parse XML are often already
27 available and very well tested.
29 The XML checking is done based on the [simgrid.dtd](http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd) Document Type
30 Definition (DTD) file.
32 If you read the DTD, you should notice the following:
33 \li The platform tag has a version attribute. The current version is <b>4</b>. This attribute might be used in the
34 provide backward compatibility.
35 \li The DTD contains definitions for both the platform description and deployment files used by SimGrid.
37 \section pf_basics Basic concepts
39 Nowadays, the Internet is composed of a bunch of independently managed
40 networks. Within each of those networks, there are entry and exit
41 points (most of the time, you can both enter and exit through the same
42 point); this allows to leave the current network and reach other
43 networks, possibly even in other locations. At the upper level, such a
44 network is called <b>Autonomous System (AS)</b>, while at the lower
45 level it is named sub-network, or LAN (local area network). They are
46 indeed autonomous: routing is defined (within the limits of his
47 network) by the administrator, and so, those networks can operate
48 without a connection to other networks. So-called gateways allow you
49 to go from one network to another, if such a (physical) connection
50 exists. Every node in one network that can be directly reached (i.e.,
51 without traversing other nodes) from another network is called a
52 gateway. Each autonomous system consists of equipment such as cables
53 (network links), routers and switches as well as computers.
55 The structure of the SimGrid platform description relies exactly on
56 the same concept as a real-life platform (see above). Every resource
57 (computers, network equipment etc.) belongs to an AS, which can be
58 defined by using the \<AS\> tag. Within an AS, the routing between its
59 elements can be defined abitrarily. There are several modes for
60 routing, and exactly one mode must be selected by specifying the
61 routing attribute in the AS tag:
64 <AS id="AS0" routing="Full">
68 Other supported values for the routing attribute can be found below, Section
71 There is also the ``<route>`` tag; this tag takes two attributes,
72 ``src`` (source) and ``dst`` (destination). Both source and
73 destination must be valid identifiers for routers (these will be
74 introduced later). Contained by the ``<route>`` are network links;
75 these links must be used in order to communicate from the source to
76 the destination specified in the tag. Hence, a route merely describes
77 how to reach a router from another router.
80 More information and (code-)examples can be found in Section \ref pf_rm.
82 An AS can also contain itself one or more AS; this allows you to model
83 the hierarchy of your platform.
85 ### Within each AS, the following types of resources exist:
87 %Resource | Documented in Section | Description
88 --------------- | --------------------- | -----------
89 AS | | Every Autonomous System (AS) may contain one or more AS.
90 host | \ref pf_host | This entity carries out the actual computation. For this reason, it contains processors (with potentially multiple cores).
91 router | \ref pf_router | In SimGrid, routers are used to provide helpful information to routing algorithms. Routers may also act as gateways, connecting several autonomous systems with each other.
92 link | \ref pf_link | In SimGrid, (network)links define a connection between two or potentially even more resources. Every link has a bandwidth and a latency and may potentially experience congestion.
93 cluster | \ref pf_cluster | In SimGrid, clusters were introduced to model large and homogenous environments. They are not really a resource by themselves - technically, they are only a shortcut, as they will internally set up all the hosts, network and routing for you, i.e., using this resource, one can easily setup thousands of hosts and links in a few lines of code. Each cluster is itself an AS.
95 As it is desirable to interconnect these resources, a routing has to
96 be defined. The AS is supposed to be Autonomous, hence this has to be
97 done at the AS level. The AS handles two different types of entities
98 (<b>host/router</b> and <b>AS</b>). However, the user is responsible
99 to define routes between those resources, otherwise entities will be
100 unconnected and therefore unreachable from other entities. Although
101 several routing algorithms are built into SimGrid (see \ref pf_rm),
102 you might encounter a case where you want to define routes manually
103 (for instance, due to specific requirements of your platform).
105 There are three tags to use:
106 \li <b>ASroute</b>: to define routes between two <b>AS</b>
107 \li <b>route</b>: to define routes between two <b>host/router</b>
108 \li <b>bypassRoute</b>: to define routes between two <b>AS</b> that
109 will bypass default routing (as specified by the ``routing`` attribute
110 supplied to ``<AS>``, see above).
112 Here is an illustration of these concepts:
114 
115 Circles represent processing units and squares represent network routers. Bold
116 lines represent communication links. AS2 models the core of a national
117 network interconnecting a small flat cluster (AS4) and a larger
118 hierarchical cluster (AS5), a subset of a LAN (AS6), and a set of peers
119 scattered around the world (AS7).
121 \section pf_pftags Resource description
123 \subsection pf_As Platform: The <AS> tag
125 The concept of an AS was already outlined above (Section \ref pf_basics);
126 recall that the AS is so important because it groups other resources (such
127 as routers/hosts) together (in fact, these resources must be contained by
130 Available attributes :
132 Attribute name | Mandatory | Values | Description
133 --------------- | --------- | ------ | -----------
134 id | yes | String | The identifier of an AS; facilitates referring to this AS. ID must be unique.
135 routing | yes | Full\| Floyd\| Dijkstra\| DijkstraCache\| None\| Vivaldi\| Cluster | See Section \ref pf_rm for details.
140 <AS id="AS0" routing="Full">
141 <host id="host1" speed="1000000000"/>
142 <host id="host2" speed="1000000000"/>
143 <link id="link1" bandwidth="125000000" latency="0.000100"/>
144 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
148 In this example, AS0 contains two hosts (host1 and host2). The route
149 between the hosts goes through link1.
151 \subsection pf_Cr Computing resources: hosts, clusters and peers.
153 \subsubsection pf_host <host/>
155 A <b>host</b> represents a computer/node card. Every host is able to execute
156 code and it can send and receive data to/from other hosts. Most importantly,
157 a host can contain more than 1 core.
161 Attribute name | Mandatory | Values | Description
162 --------------- | --------- | ------ | -----------
163 id | yes | String | The identifier of the host. facilitates referring to this AS.
164 speed | yes | double (must be > 0.0) | Computational power of every core of this host in FLOPS. Must be larger than 0.0.
165 core | no | int (Default: 1) | The number of cores of this host. If more than one core is specified, the "speed" parameter refers to every core, i.e., the total computational power is no_of_cores*speed.<br /> If 6 cores are specified, up to 6 tasks can be executed without sharing the computational power; if more than 6 tasks are executed, computational power will be shared among these tasks. <br /> <b>Warning:</b> Although functional, this model was never scientifically assessed.
166 availability_file| no | string | (Relative or absolute) filename to use as input; must contain availability traces for this host. The syntax of this file is defined below. <br /> <b>Note:</b> The filename must be specified with your system's format.
167 state_file | no | string | Same mechanism as availability_file.<br /> <b>Note:</b> The filename must be specified with your system's format.
168 coordinates | no | string | Must be provided when choosing the Vivaldi, coordinate-based routing model for the AS the host belongs to. More details can be found in the Section \ref pf_P2P_tags.
169 pstate | no | double (Default: 0.0) | FIXME: Not yet documented.
171 ### Possible children: ###
173 Tag name | Description | Documentation
174 ------------ | ----------- | -------------
175 \<mount/\> | Defines mounting points between some storage resource and the host. | \ref pf_storage_entity_mount
176 \<prop/\> | The prop tag allows you to define additional information on this host following the attribute/value schema. You may want to use it to give information to the tool you use for rendering your simulation, for example. | N/A
181 <host id="host1" speed="1000000000"/>
182 <host id="host2" speed="1000000000">
183 <prop id="color" value="blue"/>
184 <prop id="rendershape" value="square"/>
189 \anchor pf_host_dynamism
190 ### Expressing dynamism ###
192 SimGrid provides mechanisms to change a hosts' availability over
193 time, using the ``availability_file`` attribute to the ``\<host\>`` tag
194 and a separate text file whose syntax is exemplified below.
196 #### Adding a trace file ####
199 <platform version="4">
200 <host id="bob" speed="500Gf" availability_file="bob.trace" />
204 #### Example of "bob.trace" file ####
213 Let us begin to explain this example by looking at line 2. (Line 1 will become clear soon).
214 The first column describes points in time, in this case, time 0. The second column
215 describes the relative amount of power this host is able to deliver (relative
216 to the maximum performance specified in the ``\<host\>`` tag). (Clearly, the
217 second column needs to contain values that are not smaller than 0 and not larger than 1).
218 In this example, our host will deliver 500 Mflop/s at time 0, as 500 Mflop/s is the
219 maximum performance of this host. At time 11.0, it will
220 deliver half of its maximum performance, i.e., 250 Mflop/s until time 20.0 when it will
221 will start delivering 80\% of its power. In this example, this amounts to 400 Mflop/s.
223 Since the periodicity in line 1 was set to be 1.0, i.e., 1 timestep, this host will
224 continue to provide 500 Mflop/s from time 21. From time 32 it will provide 250 MFlop/s and so on.
226 ### Changing initial state ###
228 It is also possible to specify whether the host is up or down by setting the
229 ``state`` attribute to either <b>ON</b> (default value) or <b>OFF</b>.
231 #### Example: Expliciting the default value "ON" ####
234 <platform version="4">
235 <host id="bob" speed="500Gf" state="ON" />
239 If you want this host to be unavailable, simply substitute ON with OFF.
241 \anchor pf_host_churn
242 ### Expressing churn ###
244 To express the fact that a host can change state over time (as in P2P
245 systems, for instance), it is possible to use a file describing the time
246 at which the host is turned on or off. An example of the content
247 of such a file is presented below.
249 #### Adding a state file ####
252 <platform version="4">
253 <host id="bob" power="500Gf" state_file="bob.fail" />
257 #### Example of "bob.fail" file ####
265 A zero value means <b>down</b> (i.e., OFF) while a positive one means <b>up and
266 running</b> (i.e., ON). From time 0.0 to time 1.0, the host is on as usual. At time 1.0, it is
267 turned off and at time 2.0, it is turned on again until time 12 (2 plus the
268 periodicity 10). It will be turned off again at time 13.0 until time 23.0, and
272 \subsubsection pf_cluster <cluster>
274 ``<cluster />`` represents a machine-cluster. It is most commonly used
275 when one wants to define many hosts and a network quickly. Technically,
276 ``cluster`` is a meta-tag: <b>from the inner SimGrid point of
277 view, a cluster is an AS where some optimized routing is defined</b>.
278 The default inner organization of the cluster is as follow:
284 ____________|__________|_____________ backbone
286 l0| l1| l2| l97| l96 | | l99
292 Here, a set of <b>host</b>s is defined. Each of them has a <b>link</b>
293 to a central backbone (backbone is a link itself, as a link can
294 be used to represent a switch, see the switch / link section
295 below for more details about it). A <b>router</b> allows to connect a
296 <b>cluster</b> to the outside world. Internally,
297 SimGrid treats a cluster as an AS containing all hosts: the router is the default
298 gateway for the cluster.
300 There is an alternative organization, which is as follows:
314 The principle is the same, except that there is no backbone. This representation
315 can be obtained easily: just do not set the bb_* attributes.
318 Attribute name | Mandatory | Values | Description
319 --------------- | --------- | ------ | -----------
320 id | yes | string | The identifier of the cluster. Facilitates referring to this cluster.
321 prefix | yes | string | Each node of the cluster has to have a name. This name will be prefixed with this prefix.
322 suffix | yes | string | Each node of the cluster will be suffixed with this suffix
323 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.
324 speed | yes | int | Same as the ``speed`` attribute of the ``\<host\>`` tag.
325 core | no | int (default: 1) | Same as the ``core`` attribute of the ``\<host\>`` tag.
326 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
327 lat | yes | int | Latency for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
328 sharing_policy | no | string | Sharing policy for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
329 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>).
330 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>).
331 bb_sharing_policy | no | string | Sharing policy for the backbone (if any). See <b>link</b> section for syntax/details.
332 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 fullduplex 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>
333 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".
334 loopback_lat | no | int | Latency for loopback (if any). See <b>link</b> section for syntax/details. See loopback_bw for more info.
335 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.
336 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. For fat trees, refer to \ref simgrid::kernel::routing::AsClusterFatTree "AsClusterFatTree documentation". For dragonfly, refer to \ref simgrid::kernel::routing::AsClusterDragonfly "AsClusterDragonfly documentation".
339 the router name is defined as the resulting String in the following
343 router_name = prefix + clusterId + _router + suffix;
347 #### Cluster example ####
349 Consider the following two (and independent) uses of the ``cluster`` tag:
352 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
353 speed="1e9" bw="125e6" lat="5E-5"/>
355 <cluster id="my_cluster_2" prefix="c-" suffix=".me" radical="0-99"
356 speed="1e9" bw="125e6" lat="5E-5"
357 bb_bw="2.25e9" bb_lat="5E-4"/>
360 The second example creates one router and 100 machines with the following names:
362 c-my_cluster_2_router.me
370 \subsubsection pf_cabinet <cabinet>
373 This tag is only available when the routing mode of the AS
374 is set to ``Cluster``.
376 The ``<cabinet />`` tag is, like the \ref pf_cluster "<cluster>" tag,
377 a meta-tag. This means that it is simply a shortcut for creating a set of (homogenous) hosts and links quickly;
378 unsurprisingly, this tag was introduced to setup cabinets in data centers quickly. Unlike
379 <cluster>, however, the <cabinet> assumes that you create the backbone
380 and routers yourself; see our examples below.
384 Attribute name | Mandatory | Values | Description
385 --------------- | --------- | ------ | -----------
386 id | yes | string | The identifier of the cabinet. Facilitates referring to this cluster.
387 prefix | yes | string | Each node of the cabinet has to have a name. This name will be prefixed with this prefix.
388 suffix | yes | string | Each node of the cabinet will be suffixed with this suffix
389 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.
390 speed | yes | int | Same as the ``speed`` attribute of the \ref pf_host "<host>" tag.
391 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
392 lat | yes | int | Latency for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
395 Please note that as of now, it is impossible to change attributes such as,
396 amount of cores (always set to 1), the initial state of hosts/links
397 (always set to ON), the sharing policy of the links (always set to \ref pf_sharing_policy_fullduplex "FULLDUPLEX").
401 The following example was taken from ``examples/platforms/meta_cluster.xml`` and
402 shows how to use the cabinet tag.
405 <AS id="my_cluster1" routing="Cluster">
406 <cabinet id="cabinet1" prefix="host-" suffix=".cluster1"
407 speed="1Gf" bw="125MBps" lat="100us" radical="1-10"/>
408 <cabinet id="cabinet2" prefix="host-" suffix=".cluster1"
409 speed="1Gf" bw="125MBps" lat="100us" radical="11-20"/>
410 <cabinet id="cabinet3" prefix="host-" suffix=".cluster1"
411 speed="1Gf" bw="125MBps" lat="100us" radical="21-30"/>
413 <backbone id="backbone1" bandwidth="2.25GBps" latency="500us"/>
418 Please note that you must specify the \ref pf_backbone "<backbone>"
419 tag by yourself; this is not done automatically and there are no checks
420 that ensure this backbone was defined.
422 The hosts generated in the above example are named host-1.cluster, host-2.cluster1
426 \subsubsection pf_peer The <peer> tag
428 This tag represents a peer, as in Peer-to-Peer (P2P) networks. However, internally,
429 SimGrid transforms a peer into an AS (similar to Cluster). Hence, this tag
430 is virtually only a shortcut that comes with some pre-defined resources
431 and values. These are:
433 \li A tiny AS whose routing type is cluster is created
435 \li Two links: One for download and one for upload. This is
436 convenient to use and simulate stuff under the last mile model (e.g., ADSL peers).
437 \li It connects the two links to the host
438 \li It creates a router (a gateway) that serves as an entry point for this peer zone.
439 This router has coordinates.
443 Attribute name | Mandatory | Values | Description
444 --------------- | --------- | ------ | -----------
445 id | yes | string | The identifier of the peer. Facilitates referring to this peer.
446 speed | yes | int | See the description of the ``host`` tag for this attribute
447 bw_in | yes | int | Bandwidth downstream
448 bw_out | yes | int | Bandwidth upstream
449 lat | yes | double | Latency for both up- and downstream, in seconds.
450 coordinates | no | string | Coordinates of the gateway for this peer. Example value: 12.8 14.4 6.4
451 sharing_policy | no | SHARED\|FULLDUPLEX (default: FULLDUPLEX) | Sharing policy for links. See <b>link</b> description for details.
452 availability_file| no | string | Availability file for the peer. Same as host availability file. See <b>host</b> description for details.
453 state_file | no | string | State file for the peer. Same as host state file. See <b>host</b> description for details.
455 Internally, SimGrid transforms any ``\<peer/\>`` construct such as
458 coordinates="12.8 14.4 6.4"
464 into an ``\<AS\>`` (see Sections \ref pf_basics and \ref pf_As). In fact, this example of the ``\<peer/\>`` tag
465 is completely equivalent to the following declaration:
468 <AS id="as_FOO" routing="Cluster">
469 <host id="peer_FOO" speed="1.5Gf"/>
470 <link id="link_FOO_UP" bandwidth="2.25GBps" latency="500us"/>
471 <link id="link_FOO_DOWN" bandwidth="2.25GBps" latency="500us"/>
472 <router id="router_FOO" coordinates="25.5 9.4 1.4"/>
473 <host_link id="peer_FOO" up="link_FOO_UP" down="link_FOO_DOWN"/>
478 \subsection pf_ne Network equipments: links and routers
480 There are two tags at all times available to represent network entities and
481 several other tags that are available only in certain contexts.
482 1. ``<link>``: Represents a entity that has a limited bandwidth, a
483 latency, and that can be shared according to TCP way to share this
486 The concept of links in SimGrid may not be intuitive, as links are not
487 limited to connecting (exactly) two entities; in fact, you can have more than
488 two equipments connected to it. (In graph theoretical terms: A link in
489 SimGrid is not an edge, but a hyperedge)
491 2. ``<router/>``: Represents an entity that a message can be routed
492 to, but that is unable to execute any code. In SimGrid, routers have also
493 no impact on the performance: Routers do not limit any bandwidth nor
494 do they increase latency. As a matter of fact, routers are (almost) ignored
495 by the simulator when the simulation has begun.
497 3. ``<backbone/>``: This tag is only available when the containing AS is
498 used as a cluster (i.e., mode="Cluster")
501 If you want to represent an entity like a switch, you must use ``<link>`` (see section). Routers are used
502 to run some routing algorithm and determine routes (see Section \ref pf_routing for details).
504 \subsubsection pf_router <router/>
506 As said before, <b>router</b> is used only to give some information
507 for routing algorithms. So, it does not have any attributes except :
511 Attribute name | Mandatory | Values | Description
512 --------------- | --------- | ------ | -----------
513 id | yes | string | The identifier of the router to be used when referring to it.
514 coordinates | no | string | Must be provided when choosing the Vivaldi, coordinate-based routing model for the AS the router belongs to. More details can be found in the Section \ref pf_P2P_tags.
519 <router id="gw_dc1_horizdist"/>
522 \subsubsection pf_link <link/>
524 Network links can represent one-hop network connections. They are
525 characterized by their id and their bandwidth; links can (but may not) be subject
530 Attribute name | Mandatory | Values | Description
531 --------------- | --------- | ------ | -----------
532 id | yes | string | The identifier of the link to be used when referring to it.
533 bandwidth | yes | int | Maximum bandwidth for this link, given in bytes/s
534 latency | no | double (default: 0.0) | Latency for this link.
535 sharing_policy | no | \ref sharing_policy_shared "SHARED"\|\ref pf_sharing_policy_fatpipe "FATPIPE"\|\ref pf_sharing_policy_fullduplex "FULLDUPLEX" (default: SHARED) | Sharing policy for the link.
536 state | no | ON\|OFF (default: ON) | Allows you to to turn this link on or off (working / not working)
537 bandwidth_file | no | string | Allows you to use a file as input for bandwidth.
538 latency_file | no | string | Allows you to use a file as input for latency.
539 state_file | no | string | Allows you to use a file as input for states.
542 #### Possible shortcuts for ``latency`` ####
544 When using the latency attribute, you can specify the latency by using the scientific
545 notation or by using common abbreviations. For instance, the following three tags
549 <link id="LINK1" bandwidth="125000000" latency="5E-6"/>
550 <link id="LINK1" bandwidth="125000000" latency="5us"/>
551 <link id="LINK1" bandwidth="125000000" latency="0.000005"/>
554 Here, the second tag uses "us", meaning "microseconds". Other shortcuts are:
556 Name | Abbreviation | Time (in seconds)
557 ---- | ------------ | -----------------
558 Week | w | 7 * 24 * 60 * 60
559 Day | d | 24 * 60 * 60
563 Millisecond | ms | 0.001 = 10^(-3)
564 Microsecond | us | 0.000001 = 10^(-6)
565 Nanosecond | ns | 0.000000001 = 10^(-9)
566 Picosecond | ps | 0.000000000001 = 10^(-12)
568 #### Sharing policy ####
570 \anchor sharing_policy_shared
571 By default a network link is \b SHARED, i.e., if two or more data flows go
572 through a link, the bandwidth is shared fairly among all data flows. This
573 is similar to the sharing policy TCP uses.
575 \anchor pf_sharing_policy_fatpipe
576 On the other hand, if a link is defined as a \b FATPIPE,
577 each flow going through this link will be provided with the complete bandwidth,
578 i.e., no sharing occurs and the bandwidth is only limiting each flow individually.
579 Please note that this is really on a per-flow basis, not only on a per-host basis!
580 The complete bandwidth provided by this link in this mode
581 is ``number_of_flows*bandwidth``, with at most ``bandwidth`` being available per flow.
583 Using the FATPIPE mode allows to model backbones that won't affect performance
586 \anchor pf_sharing_policy_fullduplex
587 The last mode available is \b FULLDUPLEX. This means that SimGrid will
588 automatically generate two links (one carrying the suffix _UP and the other the
589 suffix _DOWN) for each ``<link>`` tag. This models situations when the direction
590 of traffic is important.
593 Transfers from one side to the other will interact similarly as
594 TCP when ACK returning packets circulate on the other direction. More
595 discussion about it is available in the description of link_ctn description.
597 In other words: The SHARED policy defines a physical limit for the bandwidth.
598 The FATPIPE mode defines a limit for each application,
599 with no upper total limit.
602 Tip: By using the FATPIPE mode, you can model big backbones that
603 won't affect performance (except latency).
608 <link id="SWITCH" bandwidth="125000000" latency="5E-5" sharing_policy="FATPIPE" />
611 #### Expressing dynamism and failures ####
613 Similar to hosts, it is possible to declare links whose state, bandwidth
614 or latency changes over time (see Section \ref pf_host_dynamism for details).
616 In the case of network links, the ``bandwidth`` and ``latency`` attributes are
617 replaced by the ``bandwidth_file`` and ``latency_file`` attributes.
618 The following XML snippet demonstrates how to use this feature in the platform
619 file. The structure of the files "link1.bw" and "link1.lat" is shown below.
622 <link id="LINK1" state_file="link1.fail" bandwidth="80000000" latency=".0001" bandwidth_file="link1.bw" latency_file="link1.lat" />
626 Even if the syntax is the same, the semantic of bandwidth and latency
627 trace files differs from that of host availability files. For bandwidth and
628 latency, the corresponding files do not
629 express availability as a fraction of the available capacity but directly in
630 bytes per seconds for the bandwidth and in seconds for the latency. This is
631 because most tools allowing to capture traces on real platforms (such as NWS)
632 express their results this way.
634 ##### Example of "link1.bw" file #####
642 In this example, the bandwidth changes repeatedly, with all changes
643 being repeated every 12 seconds.
645 At the beginning of the the simulation, the link's bandwidth is 80,000,000
646 B/s (i.e., 80 Mb/s); this value was defined in the XML snippet above.
647 After four seconds, it drops to 40 Mb/s (line 2), and climbs
648 back to 60 Mb/s after another 4 seconds (line 3). The value does not change any
649 more until the end of the period, that is, after 12 seconds have been simulated).
650 At this point, periodicity kicks in and this behavior is repeated: Seconds
651 12-16 will experience 80 Mb/s, 16-20 40 Mb/s etc.).
653 ##### Example of "link1.lat" file #####
662 In this example, the latency varies with a period of 5 seconds.
663 In the xml snippet above, the latency is initialized to be 0.0001s (100µs). This
664 value will be kept during the first second, since the latency_file contains
665 changes to this value at second one, two and three.
666 At second one, the value will be 0.001, i.e., 1ms. One second later it will
667 be adjusted to 0.01 (or 10ms) and one second later it will be set again to 1ms. The
668 value will not change until second 5, when the periodicity defined in line 1
669 kicks in. It then loops back, starting at 100µs (the initial value) for one second.
672 #### The ``<prop/>`` tag ####
674 Similar to the ``<host>`` tag, a link may also contain the ``<prop/>`` tag; see the host
675 documentation (Section \ref pf_host) for an example.
678 \subsubsection pf_backbone <backbone/>
681 This tag is <b>only available</b> when the containing AS uses the "Cluster" routing mode!
683 Using this tag, you can designate an already existing link to be a backbone.
685 Attribute name | Mandatory | Values | Description
686 --------------- | --------- | ------ | -----------
687 id | yes | string | Name of the link that is supposed to act as a backbone.
689 \subsection pf_storage Storage
692 This is a prototype version that should evolve quickly, hence this
693 is just some doc valuable only at the time of writing.
694 This section describes the storage management under SimGrid ; nowadays
695 it's only usable with MSG. It relies basically on linux-like concepts.
696 You also may want to have a look to its corresponding section in
697 @ref msg_file ; access functions are organized as a POSIX-like
700 \subsubsection pf_sto_conc Storage - Main Concepts
702 The storage facilities implemented in SimGrid help to model (and account for)
703 storage devices, such as tapes, hard-drives, CD or DVD devices etc.
704 A typical situation is depicted in the figure below:
706 \image html ./webcruft/storage_sample_scenario.png
707 \image latex ./webcruft/storage_sample_scenario.png "storage_sample_scenario" width=\textwidth
709 In this figure, two hosts called Bob and Alice are interconnected via a network
710 and each host is physically attached to a disk; it is not only possible for each host to
711 mount the disk they are attached to directly, but they can also mount disks
712 that are in a remote location. In this example, Bob mounts Alice's disk remotely
713 and accesses the storage via the network.
715 SimGrid provides 3 different entities that can be used to model setups
716 that include storage facilities:
718 Entity name | Description
719 --------------- | -----------
720 \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).
721 \ref pf_storage_entity_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.
722 \ref pf_storage_entity_mount "mount" | Must be wrapped by a \ref pf_host tag; declares which storage(s) this host has mounted and where (i.e., the mountpoint).
725 \anchor pf_storage_content_file
726 ### %Storage Content File ###
728 In order to assess exactly how much time is spent reading from the storage,
729 SimGrid needs to know what is stored on the storage device (identified by distinct (file-)name, like in a file system)
730 and what size this content has.
733 The content file is never changed by the simulation; it is parsed once
734 per simulation and kept in memory afterwards. When the content of the
735 storage changes, only the internal SimGrid data structures change.
737 \anchor pf_storage_content_file_structure
738 #### Structure of a %Storage Content File ####
740 Here is an excerpt from two storage content file; if you want to see the whole file, check
741 the file ``examples/platforms/content/storage_content.txt`` that comes with the
744 SimGrid essentially supports two different formats: UNIX-style filepaths should
745 follow the well known format:
748 /lib/libsimgrid.so.3.6.2 12710497
752 /bin/simgrid_update_xml 5018
753 /bin/graphicator 66986
754 /bin/simgrid-colorizer 2993
759 Windows filepaths, unsurprisingly, use the windows style:
762 \Windows\avastSS.scr 41664
763 \Windows\bfsvc.exe 75264
764 \Windows\bootstat.dat 67584
765 \Windows\CoreSingleLanguage.xml 31497
767 \Windows\dchcfg64.exe 335464
768 \Windows\dcmdev64.exe 93288
772 The different file formats come at a cost; in version 3.12 (and most likely
773 in later versions, too), copying files from windows-style storages to unix-style
774 storages (and vice versa) is not supported.
776 \anchor pf_storage_content_file_create
777 #### Generate a %Storage Content File ####
779 If you want to generate a storage content file based on your own filesystem (or at least a filesystem you have access to),
780 try running this command (works only on unix systems):
783 find . -type f -exec ls -1s --block=1 {} \; 2>/dev/null | awk '{ print $2 " " $1}' > ./content.txt
786 \subsubsection pf_storage_entities The Storage Entities
788 These are the entities that you can use in your platform files to include
789 storage in your model. See also the list of our \ref pf_storage_example_files "example files";
790 these might also help you to get started.
792 \anchor pf_storage_entity_storage_type
793 #### \<storage_type\> ####
795 Attribute name | Mandatory | Values | Description
796 --------------- | --------- | ------ | -----------
797 id | yes | string | Identifier of this storage_type; used when referring to it
798 model | yes | string | For reasons of future backwards compatibility only; specifies the name of the model for the storage that should be used
799 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)
800 content | yes | string | Path to a \ref pf_storage_content_file "Storage Content File" on your system. This file must exist.
801 content_type | no | ("txt_unix"\|"txt_win") | Determines which kind of filesystem you're using; make sure the filenames (stored in that file, see \ref pf_storage_content_file_structure "Storage Content File Structure"!)
803 This tag must contain some predefined model properties, specified via the <model_prop> tag. Here is a list,
804 see below for an example:
806 Property id | Mandatory | Values | Description
807 --------------- | --------- | ------ | -----------
808 Bwrite | yes | string | Bandwidth for write access; in B/s (but you can also specify e.g. "30MBps")
809 Bread | yes | string | Bandwidth for read access; in B/s (but you can also specify e.g. "30MBps")
810 Bconnexion | yes | string | Throughput (of the storage connector) in B/s.
813 A storage_type can also contain the <b><prop></b> tag. The <prop> tag allows you
814 to associate additional information to this <storage_type> and follows the
815 attribute/value schema; see the example below. You may want to use it to give information to
816 the tool you use for rendering your simulation, for example.
818 Here is a complete example for the ``storage_type`` tag:
820 <storage_type id="single_HDD" model="linear_no_lat" size="4000" content_type="txt_unix">
821 <model_prop id="Bwrite" value="30MBps" />
822 <model_prop id="Bread" value="100MBps" />
823 <model_prop id="Bconnection" value="150MBps" />
824 <prop id="Brand" value="Western Digital" />
828 \anchor pf_storage_entity_storage
829 #### <storage> ####
831 ``storage`` attributes:
833 Attribute name | Mandatory | Values | Description
834 -------------- | --------- | ------ | -----------
835 id | yes | string | Identifier of this ``storage``; used when referring to it
836 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.
837 attach | yes | string | Name of a host (see Section \ref pf_host) to which this storage is <i>physically</i> attached to (e.g., a hard drive in a computer)
838 content | no | string | When specified, overwrites the content attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
839 content_type | no | string | When specified, overwrites the content_type attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
841 Here are two examples:
844 <storage id="Disk1" typeId="single_HDD" attach="bob" />
846 <storage id="Disk2" typeId="single_SSD"
847 content="content/win_storage_content.txt"
848 content_type="txt_windows" attach="alice" />
851 The first example is straightforward: A disk is defined and called "Disk1"; it is
852 of type "single_HDD" (shown as an example of \ref pf_storage_entity_storage_type "\<storage_type\>" above) and attached
853 to a host called "bob" (the definition of this host is omitted here).
855 The second storage is called "Disk2", is still of the same type as Disk1 but
856 now specifies a new content file (so the contents will be different from Disk1)
857 and the filesystem uses the windows style; finally, it is attached to a second host,
858 called alice (which is again not defined here).
860 \anchor pf_storage_entity_mount
861 #### <mount> ####
864 | Attribute name | Mandatory | Values | Description |
865 | ---------------- | ----------- | -------- | ------------- |
866 | id | yes | string | Refers to a \ref pf_storage_entity_storage "<storage>" entity that will be mounted on that computer |
867 | name | yes | string | Path/location to/of the logical reference (mount point) of this disk
869 This tag must be enclosed by a \ref pf_host tag. It then specifies where the mountpoint of a given storage device (defined by the ``id`` attribute)
870 is; this location is specified by the ``name`` attribute.
872 Here is a simple example, taken from the file ``examples/platform/storage.xml``:
875 <storage_type id="single_SSD" model="linear_no_lat" size="500GiB">
876 <model_prop id="Bwrite" value="60MBps" />
877 <model_prop id="Bread" value="200MBps" />
878 <model_prop id="Bconnection" value="220MBps" />
881 <storage id="Disk2" typeId="single_SSD"
882 content="content/win_storage_content.txt"
883 content_type="txt_windows" attach="alice" />
884 <storage id="Disk4" typeId="single_SSD"
885 content="content/small_content.txt"
886 content_type="txt_unix" attach="denise"/>
888 <host id="alice" speed="1Gf">
889 <mount storageId="Disk2" name="c:"/>
892 <host id="denise" speed="1Gf">
893 <mount storageId="Disk2" name="c:"/>
894 <mount storageId="Disk4" name="/home"/>
898 This example is quite interesting, as the same device, called "Disk2", is mounted by
899 two hosts at the same time! Note, however, that the host called ``alice`` is actually
900 attached to this storage, as can be seen in the \ref pf_storage_entity_storage "<storage>"
901 tag. This means that ``denise`` must access this storage through the network, but SimGrid automatically takes
902 care of that for you.
904 Furthermore, this example shows that ``denise`` has mounted two storages with different
905 filesystem types (unix and windows). In general, a host can mount as many storage devices as
909 Again, the difference between ``attach`` and ``mount`` is simply that
910 an attached storage is always physically inside (or connected to) that machine;
911 for instance, a USB stick is attached to one and only one machine (where it's plugged-in)
912 but it can only be mounted on others, as mounted storage can also be a remote location.
914 ###### Example files #####
916 \verbinclude example_filelist_xmltag_mount
918 \subsubsection pf_storage_example_files Example files
920 Several examples were already discussed above; if you're interested in full examples,
921 check the the following platforms:
923 1. ``examples/platforms/storage.xml``
924 2. ``examples/platforms/remote_io.xml``
926 If you're looking for some examplary C code, you may find the source code
927 available in the directory ``examples/msg/io/`` useful.
929 \subsubsection pf_storage_examples_modelling Modelling different situations
931 The storage functionality of SimGrid is type-agnostic, that is, the implementation
932 does not presume any type of storage, such as HDDs/SSDs, RAM,
933 CD/DVD devices, USB sticks etc.
935 This allows the user to apply the simulator for a wide variety of scenarios; one
936 common scenario would be the access of remote RAM.
938 #### Modelling the access of remote RAM ####
940 How can this be achieved in SimGrid? Let's assume we have a setup where three hosts
941 (HostA, HostB, HostC) need to access remote RAM:
951 An easy way to model this scenario is to setup and define the RAM via the
952 \ref pf_storage_entity_storage "storage" and \ref pf_storage_entity_storage_type "storage type"
953 entities and attach it to a remote dummy host; then, every host can have their own links
954 to this host (modelling for instance certain scenarios, such as PCIe ...)
959 RAM - Dummy -- Host B
964 Now, if read from this storage, the host that mounts this storage
965 communicates to the dummy host which reads from RAM and
966 sends the information back.
969 \section pf_routing Routing
971 To achieve high performance, the routing tables used within SimGrid are
972 static. This means that routing between two nodes is calculated once
973 and will not change during execution. The SimGrid team chose to use this
974 approach as it is rare to have a real deficiency of a resource;
975 most of the time, a communication fails because the links experience too much
976 congestion and hence, your connection stops before the timeout or
977 because the computer designated to be the destination of that message
980 We also chose to use shortest paths algorithms in order to emulate
981 routing. Doing so is consistent with the reality: [RIP](https://en.wikipedia.org/wiki/Routing_Information_Protocol),
982 [OSPF](https://en.wikipedia.org/wiki/Open_Shortest_Path_First), [BGP](https://en.wikipedia.org/wiki/Border_Gateway_Protocol)
983 are all calculating shortest paths. They do require some time to converge, but
984 eventually, when the routing tables have stabilized, your packets will follow
987 \subsection pf_rm Routing models
989 For each AS, you must define explicitly which routing model will
990 be used. There are 3 different categories for routing models:
992 1. \ref pf_routing_model_shortest_path "Shortest-path" based models: SimGrid calculates shortest
993 paths and manages them. Behaves more or less like most real life
995 2. \ref pf_routing_model_manual "Manually-entered" route models: you have to define all routes
996 manually in the platform description file; this can become
997 tedious very quickly, as it is very verbose.
998 Consistent with some manually managed real life routing.
999 3. \ref pf_routing_model_simple "Simple/fast models": those models offer fast, low memory routing
1000 algorithms. You should consider to use this type of model if
1001 you can make some assumptions about your AS.
1002 Routing in this case is more or less ignored.
1004 \subsubsection pf_raf The router affair
1006 Using routers becomes mandatory when using shortest-path based
1007 models or when using the bindings to the ns-3 packet-level
1008 simulator instead of the native analytical network model implemented
1011 For graph-based shortest path algorithms, routers are mandatory, because these
1012 algorithms require a graph as input and so we need to have source and
1013 destination for each edge.
1015 Routers are naturally an important concept ns-3 since the
1016 way routers run the packet routing algorithms is actually simulated.
1017 SimGrid's analytical models however simply aggregate the routing time
1018 with the transfer time.
1020 So why did we incorporate routers in SimGrid? Rebuilding a graph representation
1021 only from the route information turns out to be a very difficult task, because
1022 of the missing information about how routes intersect. That is why we
1023 introduced routers, which are simply used to express these intersection points.
1024 It is important to understand that routers are only used to provide topological
1027 To express this topological information, a <b>route</b> has to be
1028 defined in order to declare which link is connected to a router.
1031 \subsubsection pf_routing_model_shortest_path Shortest-path based models
1033 The following table shows all the models that compute routes using
1034 shortest-paths algorithms are currently available in SimGrid. More detail on how
1035 to choose the best routing model is given in the Section called \"\ref pf_routing_howto_choose_wisely\".
1037 | Name | Description |
1038 | --------------------------------------------------- | -------------------------------------------------------------------------- |
1039 | \ref pf_routing_model_floyd "Floyd" | Floyd routing data. Pre-calculates all routes once |
1040 | \ref pf_routing_model_dijkstra "Dijkstra" | Dijkstra routing data. Calculates routes only when needed |
1041 | \ref pf_routing_model_dijkstracache "DijkstraCache" | Dijkstra routing data. Handles some cache for already calculated routes. |
1043 All those shortest-path models are instanciated in the same way and are
1044 completely interchangeable. Here are some examples:
1046 \anchor pf_routing_model_floyd
1051 <AS id="AS0" routing="Floyd">
1053 <cluster id="my_cluster_1" prefix="c-" suffix=""
1054 radical="0-1" speed="1000000000" bw="125000000" lat="5E-5"
1055 router_id="router1"/>
1057 <AS id="AS1" routing="None">
1058 <host id="host1" speed="1000000000"/>
1061 <link id="link1" bandwidth="100000" latency="0.01"/>
1063 <ASroute src="my_cluster_1" dst="AS1"
1066 <link_ctn id="link1"/>
1072 ASroute given at the end gives a topological information: link1 is
1073 between router1 and host1.
1075 #### Example platform files ####
1077 This is an automatically generated list of example files that use the Floyd
1078 routing model (the path is given relative to SimGrid's source directory)
1080 \verbinclude example_filelist_routing_floyd
1082 \anchor pf_routing_model_dijkstra
1085 #### Example platform files ####
1087 This is an automatically generated list of example files that use the Dijkstra
1088 routing model (the path is given relative to SimGrid's source directory)
1090 \verbinclude example_filelist_routing_dijkstra
1094 <AS id="AS_2" routing="Dijkstra">
1095 <host id="AS_2_host1" speed="1000000000"/>
1096 <host id="AS_2_host2" speed="1000000000"/>
1097 <host id="AS_2_host3" speed="1000000000"/>
1098 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1099 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1100 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1101 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1102 <router id="central_router"/>
1103 <router id="AS_2_gateway"/>
1104 <!-- routes providing topological information -->
1105 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1106 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1107 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1108 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1112 \anchor pf_routing_model_dijkstracache
1113 ### DijkstraCache ###
1115 DijkstraCache example:
1117 <AS id="AS_2" routing="DijkstraCache">
1118 <host id="AS_2_host1" speed="1000000000"/>
1120 (platform unchanged compared to upper example)
1123 #### Example platform files ####
1125 This is an automatically generated list of example files that use the DijkstraCache
1126 routing model (the path is given relative to SimGrid's source directory):
1128 Editor's note: At the time of writing, no platform file used this routing model - so
1129 if there are no example files listed here, this is likely to be correct.
1131 \verbinclude example_filelist_routing_dijkstra_cache
1133 \subsubsection pf_routing_model_manual Manually-entered route models
1135 | Name | Description |
1136 | ---------------------------------- | ------------------------------------------------------------------------------ |
1137 | \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. |
1139 \anchor pf_routing_model_full
1144 <AS id="AS0" routing="Full">
1145 <host id="host1" speed="1000000000"/>
1146 <host id="host2" speed="1000000000"/>
1147 <link id="link1" bandwidth="125000000" latency="0.000100"/>
1148 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
1152 #### Example platform files ####
1154 This is an automatically generated list of example files that use the Full
1155 routing model (the path is given relative to SimGrid's source directory):
1157 \verbinclude example_filelist_routing_full
1159 \subsubsection pf_routing_model_simple Simple/fast models
1161 | Name | Description |
1162 | ---------------------------------------- | ------------------------------------------------------------------------------ |
1163 | \ref pf_routing_model_cluster "Cluster" | This is specific to the \ref pf_cluster "<cluster/>" tag and should not be used by the user, as several assumptions are made. |
1164 | \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. |
1165 | \ref pf_routing_model_vivaldi "Vivaldi" | Perfect when you want to use coordinates. Also see the corresponding \ref pf_P2P_tags "P2P section" below. |
1167 \anchor pf_routing_model_cluster
1171 In this mode, the \ref pf_cabinet "<cabinet/>" tag is available.
1173 #### Example platform files ####
1175 This is an automatically generated list of example files that use the Cluster
1176 routing model (the path is given relative to SimGrid's source directory):
1178 \verbinclude example_filelist_routing_cluster
1180 \anchor pf_routing_model_none
1183 This model does exactly what it's name advertises: Nothing. There is no routing
1184 available within this model and if you try to communicate within the AS that
1185 uses this model, SimGrid will fail unless you have explicitly activated the
1186 \ref options_model_select_network_constant "Constant Network Model" (this model charges
1187 the same for every single communication). It should
1188 be noted, however, that you can still attach an \ref pf_routing_tag_asroute "ASroute",
1189 as is demonstrated in the example below:
1191 \verbinclude platforms/cluster_and_one_host.xml
1193 #### Example platform files ####
1195 This is an automatically generated list of example files that use the None
1196 routing model (the path is given relative to SimGrid's source directory):
1198 \verbinclude example_filelist_routing_none
1201 \anchor pf_routing_model_vivaldi
1204 For more information on how to use the [Vivaldi Coordinates](https://en.wikipedia.org/wiki/Vivaldi_coordinates),
1205 see also Section \ref pf_P2P_tags "P2P tags".
1207 For documentation on how to activate this model (as some initialization must be done
1208 in the simulator), see Section \ref options_model_network_coord "Activating Coordinate Based Routing".
1210 Note that it is possible to combine the Vivaldi routing model with other routing models;
1211 an example can be found in the file \c examples/platforms/cloud.xml. This
1212 examples models an AS using Vivaldi that contains other ASes that use different
1215 #### Example platform files ####
1217 This is an automatically generated list of example files that use the None
1218 routing model (the path is given relative to SimGrid's source directory):
1220 \verbinclude example_filelist_routing_vivaldi
1223 \subsection ps_dec Defining routes
1225 There are currently four different ways to define routes:
1227 | Name | Description |
1228 | ------------------------------------------------- | ----------------------------------------------------------------------------------- |
1229 | \ref pf_routing_tag_route "route" | Used to define route between host/router |
1230 | \ref pf_routing_tag_asroute "ASroute" | Used to define route between different AS |
1231 | \ref pf_routing_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. |
1232 | \ref pf_routing_tag_bypassasroute "bypassASroute" | Used in the same way as bypassRoute, but for AS |
1234 Basically all those tags will contain an (ordered) list of references
1235 to link that compose the route you want to define.
1237 Consider the example below:
1240 <route src="Alice" dst="Bob">
1241 <link_ctn id="link1"/>
1242 <link_ctn id="link2"/>
1243 <link_ctn id="link3"/>
1247 The route here from host Alice to Bob will be first link1, then link2,
1248 and finally link3. What about the reverse route? \ref pf_routing_tag_route "Route" and
1249 \ref pf_routing_tag_asroute "ASroute" have an optional attribute \c symmetrical, that can
1250 be either \c YES or \c NO. \c YES means that the reverse route is the same
1251 route in the inverse order, and is set to \c YES by default. Note that
1252 this is not the case for bypass*Route, as it is more probable that you
1253 want to bypass only one default route.
1255 For an \ref pf_routing_tag_asroute "ASroute", things are just slightly more complicated, as you have
1256 to give the id of the gateway which is inside the AS you want to access ...
1257 So it looks like this:
1260 <ASroute src="AS1" dst="AS2"
1261 gw_src="router1" gw_dst="router2">
1262 <link_ctn id="link1"/>
1266 gw == gateway, so when any message are trying to go from AS1 to AS2,
1267 it means that it must pass through router1 to get out of the AS, then
1268 pass through link1, and get into AS2 by being received by router2.
1269 router1 must belong to AS1 and router2 must belong to AS2.
1271 \subsubsection pf_linkctn <link_ctn/>
1273 This entity has only one purpose: Refer to an already existing
1274 \ref pf_link "<link/>" when defining a route, i.e., it
1275 can only occur as a child of \ref pf_routing_tag_route "<route/>"
1277 | Attribute name | Mandatory | Values | Description |
1278 | --------------- | --------- | ------ | ----------- |
1279 | id | yes | String | The identifier of the link that should be added to the route. |
1280 | direction | maybe | UP\|DOWN | If the link referenced by \c id has been declared as \ref pf_sharing_policy_fullduplex "FULLDUPLEX", this indicates which direction the route traverses through this link: UP or DOWN. If you don't use FULLDUPLEX, do not use this attribute or SimGrid will not find the right link.
1282 #### Example Files ####
1284 This is an automatically generated list of example files that use the \c <link_ctn/>
1285 entity (the path is given relative to SimGrid's source directory):
1287 \verbinclude example_filelist_xmltag_linkctn
1289 \subsubsection pf_routing_tag_asroute ASroute
1291 The purpose of this entity is to define a route between two ASes.
1292 This is mainly useful when you're in the \ref pf_routing_model_full "Full routing model".
1294 #### Attributes ####
1296 | Attribute name | Mandatory | Values | Description |
1297 | --------------- | --------- | ------ | ----------- |
1298 | src | yes | String | The identifier of the source AS |
1299 | dst | yes | String | See the \c src attribute |
1300 | gw_src | yes | String | The gateway that will be used within the src AS; this can be any \ref pf_host "Host" or \ref pf_router "Router" defined within the src AS. |
1301 | gw_dst | yes | String | Same as \c gw_src, but with the dst AS instead. |
1302 | symmetrical | no | YES\|NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1307 <AS id="AS0" routing="Full">
1308 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
1309 radical="0-149" speed="1000000000" bw="125000000" lat="5E-5"
1310 bb_bw="2250000000" bb_lat="5E-4"/>
1312 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
1313 radical="150-299" speed="1000000000" bw="125000000" lat="5E-5"
1314 bb_bw="2250000000" bb_lat="5E-4"/>
1316 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1318 <ASroute src="my_cluster_1" dst="my_cluster_2"
1319 gw_src="c-my_cluster_1_router.me"
1320 gw_dst="c-my_cluster_2_router.me">
1321 <link_ctn id="backbone"/>
1323 <ASroute src="my_cluster_2" dst="my_cluster_1"
1324 gw_src="c-my_cluster_2_router.me"
1325 gw_dst="c-my_cluster_1_router.me">
1326 <link_ctn id="backbone"/>
1331 \subsubsection pf_routing_tag_route route
1333 The principle is the same as for
1334 \ref pf_routing_tag_asroute "ASroute": The route contains a list of links that
1335 provide a path from \c src to \c dst. Here, \c src and \c dst can both be either a
1336 \ref pf_host "host" or \ref pf_router "router". This is mostly useful for the
1337 \ref pf_routing_model_full "Full routing model" as well as for the
1338 \ref pf_routing_model_shortest_path "shortest-paths" based models (as they require
1339 topological information).
1342 | Attribute name | Mandatory | Values | Description |
1343 | --------------- | --------- | ---------------------- | ----------- |
1344 | src | yes | String | The value given to the source's "id" attribute |
1345 | dst | yes | String | The value given to the destination's "id" attribute. |
1346 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1351 A route in the \ref pf_routing_model_full "Full routing model" could look like this:
1353 <route src="Tremblay" dst="Bourassa">
1354 <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"/>
1358 A route in the \ref pf_routing_model_shortest_path "Shortest-Path routing model" could look like this:
1360 <route src="Tremblay" dst="Bourassa">
1365 You must only have one link in your routes when you're using them to provide
1366 topological information, as the routes here are simply the edges of the
1367 (network-)graph and the employed algorithms need to know which edge connects
1368 which pair of entities.
1370 \subsubsection pf_routing_tag_bypassasroute bypassASroute
1372 As said before, once you choose
1373 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1374 define some of your routes, which will be specific. You may also want
1375 to bypass some routes defined in lower level AS at an upper stage:
1376 <b>bypassASroute</b> is the tag you're looking for. It allows to
1377 bypass routes defined between already defined between AS (if you want
1378 to bypass route for a specific host, you should just use byPassRoute).
1379 The principle is the same as ASroute : <b>bypassASroute</b> contains
1380 list of links that are in the path between src and dst.
1382 #### Attributes ####
1384 | Attribute name | Mandatory | Values | Description |
1385 | --------------- | --------- | ---------------------- | ----------- |
1386 | src | yes | String | The value given to the source AS's "id" attribute |
1387 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1388 | gw_src | yes | String | The value given to the source gateway's "id" attribute; this can be any host or router within the src AS |
1389 | gw_dst | yes | String | The value given to the destination gateway's "id" attribute; this can be any host or router within the dst AS|
1390 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1395 <bypassASRoute src="my_cluster_1" dst="my_cluster_2"
1396 gw_src="my_cluster_1_router"
1397 gw_dst="my_cluster_2_router">
1398 <link_ctn id="link_tmp"/>
1402 This example shows that link \c link_tmp (definition not displayed here) directly
1403 connects the router \c my_cluster_1_router in the source cluster to the router
1404 \c my_cluster_2_router in the destination router. Additionally, as the \c symmetrical
1405 attribute was not given, this route is presumed to be symmetrical.
1407 \subsubsection pf_routing_tag_bypassroute bypassRoute
1409 As said before, once you choose
1410 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1411 define some of your routes, which will be specific. You may also want
1412 to bypass some routes defined in lower level AS at an upper stage :
1413 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
1414 routes defined between <b>host/router</b>. The principle is the same
1415 as route : <b>bypassRoute</b> contains list of links references of
1416 links that are in the path between src and dst.
1418 #### Attributes ####
1420 | Attribute name | Mandatory | Values | Description |
1421 | --------------- | --------- | ---------------------- | ----------- |
1422 | src | yes | String | The value given to the source AS's "id" attribute |
1423 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1424 | symmetrical | no | YES \| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1429 <bypassRoute src="host_1" dst="host_2">
1430 <link_ctn id="link_tmp"/>
1434 This example shows that link \c link_tmp (definition not displayed here) directly
1435 connects host \c host_1 to host \c host_2. Additionally, as the \c symmetrical
1436 attribute was not given, this route is presumed to be symmetrical.
1438 \subsection pb_baroex Basic Routing Example
1440 Let's say you have an AS named AS_Big that contains two other AS, AS_1
1441 and AS_2. If you want to make a host (h1) from AS_1 with another one
1442 (h2) from AS_2 then you'll have to proceed as follows:
1443 \li First, you have to ensure that a route is defined from h1 to the
1444 AS_1's exit gateway and from h2 to AS_2's exit gateway.
1445 \li Then, you'll have to define a route between AS_1 to AS_2. As those
1446 AS are both resources belonging to AS_Big, then it has to be done
1447 at AS_big level. To define such a route, you have to give the
1448 source AS (AS_1), the destination AS (AS_2), and their respective
1449 gateway (as the route is effectively defined between those two
1450 entry/exit points). Elements of this route can only be elements
1451 belonging to AS_Big, so links and routers in this route should be
1452 defined inside AS_Big. If you choose some shortest-path model,
1453 this route will be computed automatically.
1455 As said before, there are mainly 2 tags for routing :
1456 \li <b>ASroute</b>: to define routes between two <b>AS</b>
1457 \li <b>route</b>: to define routes between two <b>host/router</b>
1459 As we are dealing with routes between AS, it means that those we'll
1460 have some definition at AS_Big level. Let consider AS_1 contains 1
1461 host, 1 link and one router and AS_2 3 hosts, 4 links and one router.
1462 There will be a central router, and a cross-like topology. At the end
1463 of the crosses arms, you'll find the 3 hosts and the router that will
1464 act as a gateway. We have to define routes inside those two AS. Let
1465 say that AS_1 contains full routes, and AS_2 contains some Floyd
1466 routing (as we don't want to bother with defining all routes). As
1467 we're using some shortest path algorithms to route into AS_2, we'll
1468 then have to define some <b>route</b> to gives some topological
1469 information to SimGrid. Here is a file doing it all :
1472 <AS id="AS_Big" routing="Dijkstra">
1473 <AS id="AS_1" routing="Full">
1474 <host id="AS_1_host1" speed="1000000000"/>
1475 <link id="AS_1_link" bandwidth="1250000000" latency="5E-4"/>
1476 <router id="AS_1_gateway"/>
1477 <route src="AS_1_host1" dst="AS_1_gateway">
1478 <link_ctn id="AS_1_link"/>
1481 <AS id="AS_2" routing="Floyd">
1482 <host id="AS_2_host1" speed="1000000000"/>
1483 <host id="AS_2_host2" speed="1000000000"/>
1484 <host id="AS_2_host3" speed="1000000000"/>
1485 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1486 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1487 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1488 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1489 <router id="central_router"/>
1490 <router id="AS_2_gateway"/>
1491 <!-- routes providing topological information -->
1492 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1493 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1494 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1495 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1497 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1499 <ASroute src="AS_1" dst="AS_2"
1500 gw_src="AS_1_gateway"
1501 gw_dst="AS_2_gateway">
1502 <link_ctn id="backbone"/>
1507 \section pf_other_tags Tags not (directly) describing the platform
1509 The following tags can be used inside a \<platform\> tag even if they are not
1510 directly describing the platform:
1511 \li \ref pf_config "config": it allows you to pass some configuration stuff like, for
1512 example, the network model and so on. It follows the
1513 \li \ref pf_include "include": allows you to include another file into the current one.
1515 \subsection pf_config config
1517 The only purpose of this tag is to contain the \c prop tags, as described below.
1518 These tags will then configure the options as described by Section \ref options.
1521 #### Attributes ####
1523 | Attribute name | Mandatory | Values | Description |
1524 | --------------- | --------- | ---------------------- | ----------- |
1525 | id | yes | String | The identifier of the config tag when referring to id; this is basically useless, though. |
1527 #### Possible children ####
1529 Tag name | Description | Documentation
1530 ------------ | ----------- | -------------
1531 \<prop/\> | The prop tag allows you to define different configuration options following the attribute/value schema. See the \ref options page. | N/A
1536 <?xml version='1.0'?>
1537 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1538 <platform version="4">
1539 <config id="General">
1540 <prop id="maxmin/precision" value="0.000010"></prop>
1541 <prop id="cpu/optim" value="TI"></prop>
1542 <prop id="host/model" value="compound"></prop>
1543 <prop id="network/model" value="SMPI"></prop>
1544 <prop id="path" value="~/"></prop>
1545 <prop id="smpi/bw-factor" value="65472:0.940694;15424:0.697866;9376:0.58729"></prop>
1548 <AS id="AS0" routing="Full">
1552 \subsection pf_include include
1554 Even if it can be used in other contexts, this tag was originally created
1555 to be used with \ref pf_trace. The idea was to have a file describing the
1556 platform, and another file attaching traces of a given period to the platform.
1558 The drawback is that the file chuncks that will be included do not
1559 constitute valid XML files. This may explain why this feature was never really
1560 used in practice (as far as we know). Other mechanisms, such as the ability to load
1561 several platform files one after the other, could be considered in the future.
1563 In the meanwhile, the \c include tag allows you to import other platforms into your
1564 local file. This is done with the intention to help people
1565 combine their different AS and provide new platforms. Those files
1566 should contain XML that consists of
1567 \ref pf_include "include", \ref pf_cluster "cluster", \ref pf_peer "peer", \ref pf_As "AS", \ref pf_trace "trace", \ref pf_trace "tags".
1569 Do not forget to close the tag to make it work, or you will end up with an invalid XML file.
1571 #### Attributes ####
1573 | Attribute name | Mandatory | Values | Description |
1574 | --------------- | --------- | ---------------------- | ----------- |
1575 | file | yes | String | Filename of the path you want to include with either relative or absolute path. |
1580 The following example includes two files, clusterA.xml and clusterB.xml and
1581 combines them two one platform file; all hosts, routers etc. defined in
1582 each of them will then be usable.
1585 <?xml version='1.0'?>
1586 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd">
1587 <platform version="4">
1588 <AS id="main" routing="Full">
1589 <include file="clusterA.xml"></include>
1590 <include file="clusterB.xml"></include>
1595 \subsection pf_trace trace and trace_connect
1597 Both tags are an alternate way to pass files containing information on
1598 availability, state etc. to an entity. (See also, for instance, Section \ref
1599 pf_host_churn "Churn", as described for the host entity.) Instead of referring
1600 to the file directly in the host, link, or cluster tag, you proceed by defining
1601 a trace with an id corresponding to a file, later a host/link/cluster, and
1602 finally using trace_connect you say that the file trace must be used by the
1609 <AS id="AS0" routing="Full">
1610 <host id="bob" speed="1000000000"/>
1612 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1613 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1617 The order here is important. \c trace_connect must come
1618 after the elements \c trace and \c host, as both the host
1619 and the trace definition must be known when \c trace_connect
1620 is parsed; the order of \c trace and \c host is arbitrary.
1623 #### \c trace attributes ####
1626 | Attribute name | Mandatory | Values | Description |
1627 | --------------- | --------- | ---------------------- | ----------- |
1628 | id | yes | String | Identifier of this trace; this is the name you pass on to \c trace_connect. |
1629 | 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. |
1630 | trace_periodicity | yes | String | This is the same as for \ref pf_host "hosts" (see there for details) |
1632 Here is an example of trace when no file name is provided:
1635 <trace id="myTrace" periodicity="1.0">
1642 #### \c trace_connect attributes ####
1644 | Attribute name | Mandatory | Values | Description |
1645 | --------------- | --------- | ---------------------- | ----------- |
1646 | kind | no | HOST_AVAIL\|POWER\|<br/>LINK_AVAIL\|BANDWIDTH\|LATENCY (Default: HOST_AVAIL) | Describes the kind of trace. |
1647 | trace | yes | String | Identifier of the referenced trace (specified of the trace's \c id attribute) |
1648 | element | yes | String | The identifier of the referenced entity as given by its \c id attribute |
1650 \section pf_hints Hints, tips and frequently requested features
1652 Now you should know at least the syntax and be able to create a
1653 platform by your own. However, after having ourselves wrote some platforms, there
1654 are some best practices you should pay attention to in order to
1655 produce good platform and some choices you can make in order to have
1656 faster simulations. Here's some hints and tips, then.
1658 @subsection Finding the platform example that you need
1660 Most platform files that we ship are in the @c examples/platforms
1661 folder. The good old @c grep tool can find the examples you need when
1662 wondering on a specific XML tag. Here is an example session searching
1663 for @ref pf_trace "trace_connect":
1666 % cd examples/platforms
1667 % grep -R -i -n --include="*.xml" "trace_connect" .
1668 ./two_hosts_platform_with_availability_included.xml:26:<trace_connect kind="SPEED" trace="A" element="Cpu A"/>
1669 ./two_hosts_platform_with_availability_included.xml:27:<trace_connect kind="HOST_AVAIL" trace="A_failure" element="Cpu A"/>
1670 ./two_hosts_platform_with_availability_included.xml:28:<trace_connect kind="SPEED" trace="B" element="Cpu B"/>
1671 ./two_hosts.xml:17: <trace_connect trace="Tremblay_power" element="Tremblay" kind="SPEED"/>
1674 \subsection pf_as_h AS Hierarchy
1675 The AS design allows SimGrid to go fast, because computing route is
1676 done only for the set of resources defined in this AS. If you're using
1677 only a big AS containing all resource with no AS into it and you're
1678 using Full model, then ... you'll loose all interest into it. On the
1679 other hand, designing a binary tree of AS with, at the lower level,
1680 only one host, then you'll also loose all the good AS hierarchy can
1681 give you. Remind you should always be "reasonable" in your platform
1682 definition when choosing the hierarchy. A good choice if you try to
1683 describe a real life platform is to follow the AS described in
1684 reality, since this kind of trade-off works well for real life
1687 \subsection pf_exit_as Exit AS: why and how
1688 Users that have looked at some of our platforms may have notice a
1689 non-intuitive schema ... Something like that :
1693 <AS id="AS_4" routing="Full">
1694 <AS id="exitAS_4" routing="Full">
1695 <router id="router_4"/>
1697 <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"/>
1698 <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"/>
1699 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1700 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1701 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1702 <ASroute src="cl_4_1"
1704 gw_src="c_4_1-cl_4_1_router"
1705 gw_dst="c_4_2-cl_4_2_router">
1706 <link_ctn id="4_1"/>
1707 <link_ctn id="bb_4"/>
1708 <link_ctn id="4_2"/>
1710 <ASroute src="cl_4_1"
1712 gw_src="c_4_1-cl_4_1_router"
1714 <link_ctn id="4_1"/>
1715 <link_ctn id="bb_4"/>
1717 <ASroute src="cl_4_2"
1719 gw_src="c_4_2-cl_4_2_router"
1721 <link_ctn id="4_2"/>
1722 <link_ctn id="bb_4"/>
1727 In the AS_4, you have an exitAS_4 defined, containing only one router,
1728 and routes defined to that AS from all other AS (as cluster is only a
1729 shortcut for an AS, see cluster description for details). If there was
1730 an upper AS, it would define routes to and from AS_4 with the gateway
1731 router_4. It's just because, as we did not allowed (for performances
1732 issues) to have routes from an AS to a single host/router, you have to
1733 enclose your gateway, when you have AS included in your AS, within an
1734 AS to define routes to it.
1736 \subsection pf_P2P_tags P2P or how to use coordinates
1737 SimGrid allows you to use some coordinated-based system, like vivaldi,
1738 to describe a platform. The main concept is that you have some peers
1739 that are located somewhere: this is the function of the
1740 <b>coordinates</b> of the \<peer\> or \<host\> tag. There's nothing
1741 complicated in using it, here is an example:
1744 <?xml version='1.0'?>
1745 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1746 <platform version="4">
1748 <AS id="AS0" routing="Vivaldi">
1749 <host id="100030591" coordinates="25.5 9.4 1.4" speed="1.5Gf" />
1750 <host id="100036570" coordinates="-12.7 -9.9 2.1" speed="7.3Gf" />
1752 <host id="100429957" coordinates="17.5 6.7 18.8" speed="8.3Gf" />
1757 Coordinates are then used to calculate latency (in microseconds)
1758 between two hosts by calculating the distance between the two hosts
1759 coordinates with the following formula: distance( (x1, y1, z1), (x2,
1760 y2, z2) ) = euclidian( (x1,y1), (x2,y2) ) + abs(z1) + abs(z2)
1762 In other words, we take the euclidian distance on the two first
1763 dimensions, and then add the absolute values found on the third
1764 dimension. This may seem strange, but it was found to allow better
1765 approximations of the latency matrices (see the paper describing
1768 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between AS.
1769 That is for example what is commonly done when using peers (see Section \ref pf_peer).
1771 <?xml version='1.0'?>
1772 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1773 <platform version="4">
1775 <AS id="AS0" routing="Vivaldi">
1776 <peer id="peer-0" coordinates="173.0 96.8 0.1" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1777 <peer id="peer-1" coordinates="247.0 57.3 0.6" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1778 <peer id="peer-2" coordinates="243.4 58.8 1.4" speed="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1782 In such a case though, we connect the AS created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1783 This means that to route between AS1 and AS2, it will use the coordinates of router_AS1 and router_AS2.
1784 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1785 You may have noted that conveniently, a peer named FOO defines an AS named FOO and a router named router_FOO, which is why it works seamlessly with the <b>peer</b> tag.
1788 \subsection pf_routing_howto_choose_wisely Choosing wisely the routing model to use
1791 Choosing wisely the routing model to use can significantly fasten your
1792 simulation/save your time when writing the platform/save tremendous
1793 disk space. Here is the list of available model and their
1794 characteristics (lookup : time to resolve a route):
1796 \li <b>Full</b>: Full routing data (fast, large memory requirements,
1798 \li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1799 lookup, lesser memory requirements, shortest path routing only).
1800 Calculates all routes at once at the beginning.
1801 \li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1802 lookup, small memory requirements, shortest path routing only).
1803 Calculates a route when necessary.
1804 \li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1805 fast lookup, small memory requirements, shortest path routing
1806 only). Same as Dijkstra, except it handles a cache for latest used
1808 \li <b>None</b>: No routing (usable with Constant network only).
1809 Defines that there is no routes, so if you try to determine a
1810 route without constant network within this AS, SimGrid will raise
1812 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1813 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1816 \subsection pf_switch I want to describe a switch but there is no switch tag!
1818 Actually we did not include switch tag. But when you're trying to
1819 simulate a switch, assuming
1820 fluid bandwidth models are used (which SimGrid uses by default unless
1821 ns-3 or constant network models are activated), the limiting factor is
1822 switch backplane bandwidth. So, essentially, at least from
1823 the simulation perspective, a switch is similar to a
1824 link: some device that is traversed by flows and with some latency and
1825 so,e maximum bandwidth. Thus, you can simply simulate a switch as a
1827 can be connected to this "switch", which is then included in routes just
1831 \subsection pf_multicabinets I want to describe multi-cabinets clusters!
1833 You have several possibilities, as usual when modeling things. If your
1834 cabinets are homogeneous and the intercabinet network negligible for
1835 your study, you should just create a larger cluster with all hosts at
1838 In the rare case where your hosts are not homogeneous between the
1839 cabinets, you can create your cluster completely manually. For that,
1840 create an As using the Cluster routing, and then use one
1841 <cabinet> for each cabinet. This cabinet tag can only be used an
1842 As using the Cluster routing schema, and creating
1844 Be warned that creating a cluster manually from the XML with
1845 <cabinet>, <backbone> and friends is rather tedious. The
1846 easiest way to retrieve some control of your model without diving into
1847 the <cluster> internals is certainly to create one separate
1848 <cluster> per cabinet and interconnect them together. This is
1849 what we did in the G5K example platform for the Graphen cluster.
1851 \subsection pf_platform_multipath I want to express multipath routing in platform files!
1853 It is unfortunately impossible to express the fact that there is more
1854 than one routing path between two given hosts. Let's consider the
1855 following platform file:
1858 <route src="A" dst="B">
1861 <route src="B" dst="C">
1864 <route src="A" dst="C">
1869 Although it is perfectly valid, it does not mean that data traveling
1870 from A to C can either go directly (using link 3) or through B (using
1871 links 1 and 2). It simply means that the routing on the graph is not
1872 trivial, and that data do not following the shortest path in number of
1873 hops on this graph. Another way to say it is that there is no implicit
1874 in these routing descriptions. The system will only use the routes you
1875 declare (such as <route src="A" dst="C"><link_ctn
1876 id="3"/></route>), without trying to build new routes by aggregating
1879 You are also free to declare platform where the routing is not
1880 symmetrical. For example, add the following to the previous file:
1883 <route src="C" dst="A">
1889 This makes sure that data from C to A go through B where data from A
1890 to C go directly. Don't worry about realism of such settings since
1891 we've seen ways more weird situation in real settings (in fact, that's
1892 the realism of very regular platforms which is questionable, but
1893 that's another story).
