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.
44 At the upper level, such a network is called
45 <b>Autonomous System (AS)</b>, while at the lower level it is named
46 sub-network, or LAN (local area network).
47 They are indeed autonomous: routing is defined
48 (within the limits of his network) by the administrator, and so, those
49 networks can operate without a connection to other
50 networks. So-called gateways allow you to go from one network to
51 another, if such a (physical) connection exists. Every node in one network
52 that can be directly reached (i.e., without traversing other nodes) from
53 another network is called a gateway.
54 Each autonomous system consists of equipment such as cables (network links),
55 routers and switches as well as computers.
57 The structure of the SimGrid platform description relies exactly on the same
58 concept as a real-life platform (see above). Every resource (computers,
59 network equipment etc.) belongs to an AS, which can be defined by using the
60 \<AS\> tag. Within an AS, the routing between its elements can be defined
61 abitrarily. There are several modes for routing, and exactly one mode must be
62 selected by specifying the routing attribute in the AS tag:
65 <AS id="AS0" routing="Full">
69 Other supported values for the routing attribute can be found below, Section
72 There is also the ``<route>`` tag; this tag takes two attributes, ``src`` (source)
73 and ``dst`` (destination). Both source and destination must be valid identifiers
74 for routers (these will be introduced later). Contained by the ``<route>`` are
75 network links; these links must be used in order to communicate from the source
76 to 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
83 model 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 be
96 defined. The AS is supposed to be Autonomous, hence this has to be done at the
97 AS level. The AS handles two different types of entities (<b>host/router</b>
98 and <b>AS</b>). However, the user is responsible to define routes between those resources,
99 otherwise entities will be unconnected and therefore unreachable from other
100 entities. Although several routing algorithms are built into SimGrid (see
101 \ref pf_rm), you might encounter a case where you want to define routes
102 manually (for instance, due to specific requirements of your platform).
104 There are three tags to use:
105 \li <b>ASroute</b>: to define routes between two <b>AS</b>
106 \li <b>route</b>: to define routes between two <b>host/router</b>
107 \li <b>bypassRoute</b>: to define routes between two <b>AS</b> that
108 will bypass default routing (as specified by the ``routing`` attribute
109 supplied to ``<AS>``, see above).
111 Here is an illustration of these concepts:
113 
114 Circles represent processing units and squares represent network routers. Bold
115 lines represent communication links. AS2 models the core of a national
116 network interconnecting a small flat cluster (AS4) and a larger
117 hierarchical cluster (AS5), a subset of a LAN (AS6), and a set of peers
118 scattered around the world (AS7).
120 \section pf_pftags Resource description
122 \subsection pf_As Platform: The <AS> tag
124 The concept of an AS was already outlined above (Section \ref pf_basics);
125 recall that the AS is so important because it groups other resources (such
126 as routers/hosts) together (in fact, these resources must be contained by
129 Available attributes :
131 Attribute name | Mandatory | Values | Description
132 --------------- | --------- | ------ | -----------
133 id | yes | String | The identifier of an AS; facilitates referring to this AS. ID must be unique.
134 routing | yes | Full\| Floyd\| Dijkstra\| DijkstraCache\| None\| Vivaldi\| Cluster | See Section \ref pf_rm for details.
139 <AS id="AS0" routing="Full">
140 <host id="host1" power="1000000000"/>
141 <host id="host2" power="1000000000"/>
142 <link id="link1" bandwidth="125000000" latency="0.000100"/>
143 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
147 In this example, AS0 contains two hosts (host1 and host2). The route
148 between the hosts goes through link1.
150 \subsection pf_Cr Computing resources: hosts, clusters and peers.
152 \subsubsection pf_host <host/>
154 A <b>host</b> represents a computer/node card. Every host is able to execute
155 code and it can send and receive data to/from other hosts. Most importantly,
156 a host can contain more than 1 core.
160 Attribute name | Mandatory | Values | Description
161 --------------- | --------- | ------ | -----------
162 id | yes | String | The identifier of the host. facilitates referring to this AS.
163 speed | yes | double (must be > 0.0) | Computational power of every core of this host in FLOPS. Must be larger than 0.0.
164 core | no | int (Default: 1) | The number of cores of this host. If more than one core is specified, the "power" parameter refers to every core, i.e., the total computational power is no_of_cores*power.<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.
165 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.
166 state_file | no | string | Same mechanism as availability_file.<br /> <b>Note:</b> The filename must be specified with your system's format.
167 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.
168 pstate | no | double (Default: 0.0) | FIXME: Not yet documented.
170 ### Possible children: ###
172 Tag name | Description | Documentation
173 ------------ | ----------- | -------------
174 \<mount/\> | Defines mounting points between some storage resource and the host. | \ref pf_storage_entity_mount
175 \<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
180 <host id="host1" power="1000000000"/>
181 <host id="host2" power="1000000000">
182 <prop id="color" value="blue"/>
183 <prop id="rendershape" value="square"/>
188 \anchor pf_host_dynamism
189 ### Expressing dynamism ###
191 SimGrid provides mechanisms to change a hosts' availability over
192 time, using the ``availability_file`` attribute to the ``\<host\>`` tag
193 and a separate text file whose syntax is exemplified below.
195 #### Adding a trace file ####
198 <platform version="4">
199 <host id="bob" power="500Gf" availability_file="bob.trace" />
203 #### Example of "bob.trace" file ####
212 Let us begin to explain this example by looking at line 2. (Line 1 will become clear soon).
213 The first column describes points in time, in this case, time 0. The second column
214 describes the relative amount of power this host is able to deliver (relative
215 to the maximum performance specified in the ``\<host\>`` tag). (Clearly, the
216 second column needs to contain values that are not smaller than 0 and not larger than 1).
217 In this example, our host will deliver 500 Mflop/s at time 0, as 500 Mflop/s is the
218 maximum performance of this host. At time 11.0, it will
219 deliver half of its maximum performance, i.e., 250 Mflop/s until time 20.0 when it will
220 will start delivering 80\% of its power. In this example, this amounts to 400 Mflop/s.
222 Since the periodicity in line 1 was set to be 1.0, i.e., 1 timestep, this host will
223 continue to provide 500 Mflop/s from time 21. From time 32 it will provide 250 MFlop/s and so on.
225 ### Changing initial state ###
227 It is also possible to specify whether the host is up or down by setting the
228 ``state`` attribute to either <b>ON</b> (default value) or <b>OFF</b>.
230 #### Example: Expliciting the default value "ON" ####
233 <platform version="4">
234 <host id="bob" power="500Gf" state="ON" />
238 If you want this host to be unavailable, simply substitute ON with OFF.
240 \anchor pf_host_churn
241 ### Expressing churn ###
243 To express the fact that a host can change state over time (as in P2P
244 systems, for instance), it is possible to use a file describing the time
245 at which the host is turned on or off. An example of the content
246 of such a file is presented below.
248 #### Adding a state file ####
251 <platform version="4">
252 <host id="bob" power="500Gf" state_file="bob.fail" />
256 #### Example of "bob.fail" file ####
264 A zero value means <b>down</b> (i.e., OFF) while a positive one means <b>up and
265 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
266 turned off and at time 2.0, it is turned on again until time 12 (2 plus the
267 periodicity 10). It will be turned off again at time 13.0 until time 23.0, and
271 \subsubsection pf_cluster <cluster>
273 ``<cluster />`` represents a machine-cluster. It is most commonly used
274 when one wants to define many hosts and a network quickly. Technically,
275 ``cluster`` is a meta-tag: <b>from the inner SimGrid point of
276 view, a cluster is an AS where some optimized routing is defined</b>.
277 The default inner organization of the cluster is as follow:
283 ____________|__________|_____________ backbone
285 l0| l1| l2| l97| l96 | | l99
291 Here, a set of <b>host</b>s is defined. Each of them has a <b>link</b>
292 to a central backbone (backbone is a link itself, as a link can
293 be used to represent a switch, see the switch / link section
294 below for more details about it). A <b>router</b> allows to connect a
295 <b>cluster</b> to the outside world. Internally,
296 SimGrid treats a cluster as an AS containing all hosts: the router is the default
297 gateway for the cluster.
299 There is an alternative organization, which is as follows:
313 The principle is the same, except that there is no backbone. This representation
314 can be obtained easily: just do not set the bb_* attributes.
317 Attribute name | Mandatory | Values | Description
318 --------------- | --------- | ------ | -----------
319 id | yes | string | The identifier of the cluster. Facilitates referring to this cluster.
320 prefix | yes | string | Each node of the cluster has to have a name. This name will be prefixed with this prefix.
321 suffix | yes | string | Each node of the cluster will be suffixed with this suffix
322 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.
323 power | yes | int | Same as the ``power`` attribute of the ``\<host\>`` tag.
324 core | no | int (default: 1) | Same as the ``core`` attribute of the ``\<host\>`` tag.
325 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
326 lat | yes | int | Latency for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
327 sharing_policy | no | string | Sharing policy for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
328 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>).
329 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>).
330 bb_sharing_policy | no | string | Sharing policy for the backbone (if any). See <b>link</b> section for syntax/details.
331 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>
332 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".
333 loopback_lat | no | int | Latency for loopback (if any). See <b>link</b> section for syntax/details. See loopback_bw for more info.
334 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.
335 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::surf::AsClusterFatTree "AsClusterFatTree documentation". For dragonfly, refer to \ref simgrid::surf::AsClusterDragonfly "AsClusterDragonfly documentation".
338 the router name is defined as the resulting String in the following
342 router_name = prefix + clusterId + _router + suffix;
346 #### Cluster example ####
348 Consider the following two (and independent) uses of the ``cluster`` tag:
351 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
352 power="1e9" bw="125e6" lat="5E-5"/>
354 <cluster id="my_cluster_2" prefix="c-" suffix=".me" radical="0-99"
355 power="1e9" bw="125e6" lat="5E-5"
356 bb_bw="2.25e9" bb_lat="5E-4"/>
359 The second example creates one router and 100 machines with the following names:
361 c-my_cluster_2_router.me
369 \subsubsection pf_cabinet <cabinet>
372 This tag is only available when the routing mode of the AS
373 is set to ``Cluster``.
375 The ``<cabinet />`` tag is, like the \ref pf_cluster "<cluster>" tag,
376 a meta-tag. This means that it is simply a shortcut for creating a set of (homogenous) hosts and links quickly;
377 unsurprisingly, this tag was introduced to setup cabinets in data centers quickly. Unlike
378 <cluster>, however, the <cabinet> assumes that you create the backbone
379 and routers yourself; see our examples below.
383 Attribute name | Mandatory | Values | Description
384 --------------- | --------- | ------ | -----------
385 id | yes | string | The identifier of the cabinet. Facilitates referring to this cluster.
386 prefix | yes | string | Each node of the cabinet has to have a name. This name will be prefixed with this prefix.
387 suffix | yes | string | Each node of the cabinet will be suffixed with this suffix
388 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.
389 power | yes | int | Same as the ``power`` attribute of the \ref pf_host "<host>" tag.
390 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
391 lat | yes | int | Latency for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
394 Please note that as of now, it is impossible to change attributes such as,
395 amount of cores (always set to 1), the initial state of hosts/links
396 (always set to ON), the sharing policy of the links (always set to \ref pf_sharing_policy_fullduplex "FULLDUPLEX").
400 The following example was taken from ``examples/platforms/meta_cluster.xml`` and
401 shows how to use the cabinet tag.
404 <AS id="my_cluster1" routing="Cluster">
405 <cabinet id="cabinet1" prefix="host-" suffix=".cluster1"
406 power="1Gf" bw="125MBps" lat="100us" radical="1-10"/>
407 <cabinet id="cabinet2" prefix="host-" suffix=".cluster1"
408 power="1Gf" bw="125MBps" lat="100us" radical="11-20"/>
409 <cabinet id="cabinet3" prefix="host-" suffix=".cluster1"
410 power="1Gf" bw="125MBps" lat="100us" radical="21-30"/>
412 <backbone id="backbone1" bandwidth="2.25GBps" latency="500us"/>
417 Please note that you must specify the \ref pf_backbone "<backbone>"
418 tag by yourself; this is not done automatically and there are no checks
419 that ensure this backbone was defined.
421 The hosts generated in the above example are named host-1.cluster, host-2.cluster1
425 \subsubsection pf_peer The <peer> tag
427 This tag represents a peer, as in Peer-to-Peer (P2P) networks. However, internally,
428 SimGrid transforms a peer into an AS (similar to Cluster). Hence, this tag
429 is virtually only a shortcut that comes with some pre-defined resources
430 and values. These are:
432 \li A tiny AS whose routing type is cluster is created
434 \li Two links: One for download and one for upload. This is
435 convenient to use and simulate stuff under the last mile model (e.g., ADSL peers).
436 \li It connects the two links to the host
437 \li It creates a router (a gateway) that serves as an entry point for this peer zone.
438 This router has coordinates.
442 Attribute name | Mandatory | Values | Description
443 --------------- | --------- | ------ | -----------
444 id | yes | string | The identifier of the peer. Facilitates referring to this peer.
445 power | yes | int | See the description of the ``host`` tag for this attribute
446 bw_in | yes | int | Bandwidth downstream
447 bw_out | yes | int | Bandwidth upstream
448 lat | yes | double | Latency for both up- and downstream, in seconds.
449 coordinates | no | string | Coordinates of the gateway for this peer. Example value: 12.8 14.4 6.4
450 sharing_policy | no | SHARED\|FULLDUPLEX (default: FULLDUPLEX) | Sharing policy for links. See <b>link</b> description for details.
451 availability_file| no | string | Availability file for the peer. Same as host availability file. See <b>host</b> description for details.
452 state_file | no | string | State file for the peer. Same as host state file. See <b>host</b> description for details.
454 Internally, SimGrid transforms any ``\<peer/\>`` construct such as
457 coordinates="12.8 14.4 6.4"
463 into an ``\<AS\>`` (see Sections \ref pf_basics and \ref pf_As). In fact, this example of the ``\<peer/\>`` tag
464 is completely equivalent to the following declaration:
467 <AS id="as_FOO" routing="Cluster">
468 <host id="peer_FOO" power="1.5Gf"/>
469 <link id="link_FOO_UP" bandwidth="2.25GBps" latency="500us"/>
470 <link id="link_FOO_DOWN" bandwidth="2.25GBps" latency="500us"/>
471 <router id="router_FOO" coordinates="25.5 9.4 1.4"/>
472 <host_link id="peer_FOO" up="link_FOO_UP" down="link_FOO_DOWN"/>
477 \subsection pf_ne Network equipments: links and routers
479 There are two tags at all times available to represent network entities and
480 several other tags that are available only in certain contexts.
481 1. ``<link>``: Represents a entity that has a limited bandwidth, a
482 latency, and that can be shared according to TCP way to share this
485 The concept of links in SimGrid may not be intuitive, as links are not
486 limited to connecting (exactly) two entities; in fact, you can have more than
487 two equipments connected to it. (In graph theoretical terms: A link in
488 SimGrid is not an edge, but a hyperedge)
490 2. ``<router/>``: Represents an entity that a message can be routed
491 to, but that is unable to execute any code. In SimGrid, routers have also
492 no impact on the performance: Routers do not limit any bandwidth nor
493 do they increase latency. As a matter of fact, routers are (almost) ignored
494 by the simulator when the simulation has begun.
496 3. ``<backbone/>``: This tag is only available when the containing AS is
497 used as a cluster (i.e., mode="Cluster")
500 If you want to represent an entity like a switch, you must use ``<link>`` (see section). Routers are used
501 to run some routing algorithm and determine routes (see Section \ref pf_routing for details).
503 \subsubsection pf_router <router/>
505 %As said before, <b>router</b> is used only to give some information
506 for routing algorithms. So, it does not have any attributes except :
510 Attribute name | Mandatory | Values | Description
511 --------------- | --------- | ------ | -----------
512 id | yes | string | The identifier of the router to be used when referring to it.
513 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.
518 <router id="gw_dc1_horizdist"/>
521 \subsubsection pf_link <link/>
523 Network links can represent one-hop network connections. They are
524 characterized by their id and their bandwidth; links can (but may not) be subject
529 Attribute name | Mandatory | Values | Description
530 --------------- | --------- | ------ | -----------
531 id | yes | string | The identifier of the link to be used when referring to it.
532 bandwidth | yes | int | Maximum bandwidth for this link, given in bytes/s
533 latency | no | double (default: 0.0) | Latency for this link.
534 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.
535 state | no | ON\|OFF (default: ON) | Allows you to to turn this link on or off (working / not working)
536 bandwidth_file | no | string | Allows you to use a file as input for bandwidth.
537 latency_file | no | string | Allows you to use a file as input for latency.
538 state_file | no | string | Allows you to use a file as input for states.
541 #### Possible shortcuts for ``latency`` ####
543 When using the latency attribute, you can specify the latency by using the scientific
544 notation or by using common abbreviations. For instance, the following three tags
548 <link id="LINK1" bandwidth="125000000" latency="5E-6"/>
549 <link id="LINK1" bandwidth="125000000" latency="5us"/>
550 <link id="LINK1" bandwidth="125000000" latency="0.000005"/>
553 Here, the second tag uses "us", meaning "microseconds". Other shortcuts are:
555 Name | Abbreviation | Time (in seconds)
556 ---- | ------------ | -----------------
557 Week | w | 7 * 24 * 60 * 60
558 Day | d | 24 * 60 * 60
562 Millisecond | ms | 0.001 = 10^(-3)
563 Microsecond | us | 0.000001 = 10^(-6)
564 Nanosecond | ns | 0.000000001 = 10^(-9)
565 Picosecond | ps | 0.000000000001 = 10^(-12)
567 #### Sharing policy ####
569 \anchor sharing_policy_shared
570 By default a network link is \b SHARED, i.e., if two or more data flows go
571 through a link, the bandwidth is shared fairly among all data flows. This
572 is similar to the sharing policy TCP uses.
574 \anchor pf_sharing_policy_fatpipe
575 On the other hand, if a link is defined as a \b FATPIPE,
576 each flow going through this link will be provided with the complete bandwidth,
577 i.e., no sharing occurs and the bandwidth is only limiting each flow individually.
578 Please note that this is really on a per-flow basis, not only on a per-host basis!
579 The complete bandwidth provided by this link in this mode
580 is ``number_of_flows*bandwidth``, with at most ``bandwidth`` being available per flow.
582 Using the FATPIPE mode allows to model backbones that won't affect performance
585 \anchor pf_sharing_policy_fullduplex
586 The last mode available is \b FULLDUPLEX. This means that SimGrid will
587 automatically generate two links (one carrying the suffix _UP and the other the
588 suffix _DOWN) for each ``<link>`` tag. This models situations when the direction
589 of traffic is important.
592 Transfers from one side to the other will interact similarly as
593 TCP when ACK returning packets circulate on the other direction. More
594 discussion about it is available in the description of link_ctn description.
596 In other words: The SHARED policy defines a physical limit for the bandwidth.
597 The FATPIPE mode defines a limit for each application,
598 with no upper total limit.
601 Tip: By using the FATPIPE mode, you can model big backbones that
602 won't affect performance (except latency).
607 <link id="SWITCH" bandwidth="125000000" latency="5E-5" sharing_policy="FATPIPE" />
610 #### Expressing dynamism and failures ####
612 Similar to hosts, it is possible to declare links whose state, bandwidth
613 or latency changes over time (see Section \ref pf_host_dynamism for details).
615 In the case of network links, the ``bandwidth`` and ``latency`` attributes are
616 replaced by the ``bandwidth_file`` and ``latency_file`` attributes.
617 The following XML snippet demonstrates how to use this feature in the platform
618 file. The structure of the files "link1.bw" and "link1.lat" is shown below.
621 <link id="LINK1" state_file="link1.fail" bandwidth="80000000" latency=".0001" bandwidth_file="link1.bw" latency_file="link1.lat" />
625 Even if the syntax is the same, the semantic of bandwidth and latency
626 trace files differs from that of host availability files. For bandwidth and
627 latency, the corresponding files do not
628 express availability as a fraction of the available capacity but directly in
629 bytes per seconds for the bandwidth and in seconds for the latency. This is
630 because most tools allowing to capture traces on real platforms (such as NWS)
631 express their results this way.
633 ##### Example of "link1.bw" file #####
641 In this example, the bandwidth changes repeatedly, with all changes
642 being repeated every 12 seconds.
644 At the beginning of the the simulation, the link's bandwidth is 80,000,000
645 B/s (i.e., 80 Mb/s); this value was defined in the XML snippet above.
646 After four seconds, it drops to 40 Mb/s (line 2), and climbs
647 back to 60 Mb/s after another 4 seconds (line 3). The value does not change any
648 more until the end of the period, that is, after 12 seconds have been simulated).
649 At this point, periodicity kicks in and this behavior is repeated: Seconds
650 12-16 will experience 80 Mb/s, 16-20 40 Mb/s etc.).
652 ##### Example of "link1.lat" file #####
661 In this example, the latency varies with a period of 5 seconds.
662 In the xml snippet above, the latency is initialized to be 0.0001s (100µs). This
663 value will be kept during the first second, since the latency_file contains
664 changes to this value at second one, two and three.
665 At second one, the value will be 0.001, i.e., 1ms. One second later it will
666 be adjusted to 0.01 (or 10ms) and one second later it will be set again to 1ms. The
667 value will not change until second 5, when the periodicity defined in line 1
668 kicks in. It then loops back, starting at 100µs (the initial value) for one second.
671 #### The ``<prop/>`` tag ####
673 Similar to the ``<host>`` tag, a link may also contain the ``<prop/>`` tag; see the host
674 documentation (Section \ref pf_host) for an example.
677 \subsubsection pf_backbone <backbone/>
680 This tag is <b>only available</b> when the containing AS uses the "Cluster" routing mode!
682 Using this tag, you can designate an already existing link to be a backbone.
684 Attribute name | Mandatory | Values | Description
685 --------------- | --------- | ------ | -----------
686 id | yes | string | Name of the link that is supposed to act as a backbone.
688 \subsection pf_storage Storage
691 This is a prototype version that should evolve quickly, hence this
692 is just some doc valuable only at the time of writing.
693 This section describes the storage management under SimGrid ; nowadays
694 it's only usable with MSG. It relies basically on linux-like concepts.
695 You also may want to have a look to its corresponding section in
696 @ref msg_file ; access functions are organized as a POSIX-like
699 \subsubsection pf_sto_conc Storage - Main Concepts
701 The storage facilities implemented in SimGrid help to model (and account for)
702 storage devices, such as tapes, hard-drives, CD or DVD devices etc.
703 A typical situation is depicted in the figure below:
705 \image html ./webcruft/storage_sample_scenario.png
706 \image latex ./webcruft/storage_sample_scenario.png "storage_sample_scenario" width=\textwidth
708 In this figure, two hosts called Bob and Alice are interconnected via a network
709 and each host is physically attached to a disk; it is not only possible for each host to
710 mount the disk they are attached to directly, but they can also mount disks
711 that are in a remote location. In this example, Bob mounts Alice's disk remotely
712 and accesses the storage via the network.
714 SimGrid provides 3 different entities that can be used to model setups
715 that include storage facilities:
717 Entity name | Description
718 --------------- | -----------
719 \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).
720 \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.
721 \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).
724 \anchor pf_storage_content_file
725 ### %Storage Content File ###
727 In order to assess exactly how much time is spent reading from the storage,
728 SimGrid needs to know what is stored on the storage device (identified by distinct (file-)name, like in a file system)
729 and what size this content has.
732 The content file is never changed by the simulation; it is parsed once
733 per simulation and kept in memory afterwards. When the content of the
734 storage changes, only the internal SimGrid data structures change.
736 \anchor pf_storage_content_file_structure
737 #### Structure of a %Storage Content File ####
739 Here is an excerpt from two storage content file; if you want to see the whole file, check
740 the file ``examples/platforms/content/storage_content.txt`` that comes with the
743 SimGrid essentially supports two different formats: UNIX-style filepaths should
744 follow the well known format:
747 /lib/libsimgrid.so.3.6.2 12710497
751 /bin/simgrid_update_xml 5018
752 /bin/graphicator 66986
753 /bin/simgrid-colorizer 2993
758 Windows filepaths, unsurprisingly, use the windows style:
761 \Windows\avastSS.scr 41664
762 \Windows\bfsvc.exe 75264
763 \Windows\bootstat.dat 67584
764 \Windows\CoreSingleLanguage.xml 31497
766 \Windows\dchcfg64.exe 335464
767 \Windows\dcmdev64.exe 93288
771 The different file formats come at a cost; in version 3.12 (and most likely
772 in later versions, too), copying files from windows-style storages to unix-style
773 storages (and vice versa) is not supported.
775 \anchor pf_storage_content_file_create
776 #### Generate a %Storage Content File ####
778 If you want to generate a storage content file based on your own filesystem (or at least a filesystem you have access to),
779 try running this command (works only on unix systems):
782 find . -type f -exec ls -1s --block=1 {} \; 2>/dev/null | awk '{ print $2 " " $1}' > ./content.txt
785 \subsubsection pf_storage_entities The Storage Entities
787 These are the entities that you can use in your platform files to include
788 storage in your model. See also the list of our \ref pf_storage_example_files "example files";
789 these might also help you to get started.
791 \anchor pf_storage_entity_storage_type
792 #### \<storage_type\> ####
794 Attribute name | Mandatory | Values | Description
795 --------------- | --------- | ------ | -----------
796 id | yes | string | Identifier of this storage_type; used when referring to it
797 model | yes | string | For reasons of future backwards compatibility only; specifies the name of the model for the storage that should be used
798 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)
799 content | yes | string | Path to a \ref pf_storage_content_file "Storage Content File" on your system. This file must exist.
800 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"!)
802 This tag must contain some predefined model properties, specified via the <model_prop> tag. Here is a list,
803 see below for an example:
805 Property id | Mandatory | Values | Description
806 --------------- | --------- | ------ | -----------
807 Bwrite | yes | string | Bandwidth for write access; in B/s (but you can also specify e.g. "30MBps")
808 Bread | yes | string | Bandwidth for read access; in B/s (but you can also specify e.g. "30MBps")
809 Bconnexion | yes | string | Throughput (of the storage connector) in B/s.
812 A storage_type can also contain the <b><prop></b> tag. The <prop> tag allows you
813 to associate additional information to this <storage_type> and follows the
814 attribute/value schema; see the example below. You may want to use it to give information to
815 the tool you use for rendering your simulation, for example.
817 Here is a complete example for the ``storage_type`` tag:
819 <storage_type id="single_HDD" model="linear_no_lat" size="4000" content_type="txt_unix">
820 <model_prop id="Bwrite" value="30MBps" />
821 <model_prop id="Bread" value="100MBps" />
822 <model_prop id="Bconnection" value="150MBps" />
823 <prop id="Brand" value="Western Digital" />
827 \anchor pf_storage_entity_storage
828 #### <storage> ####
830 ``storage`` attributes:
832 Attribute name | Mandatory | Values | Description
833 -------------- | --------- | ------ | -----------
834 id | yes | string | Identifier of this ``storage``; used when referring to it
835 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.
836 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)
837 content | no | string | When specified, overwrites the content attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
838 content_type | no | string | When specified, overwrites the content_type attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
840 Here are two examples:
843 <storage id="Disk1" typeId="single_HDD" attach="bob" />
845 <storage id="Disk2" typeId="single_SSD"
846 content="content/win_storage_content.txt"
847 content_type="txt_windows" attach="alice" />
850 The first example is straightforward: A disk is defined and called "Disk1"; it is
851 of type "single_HDD" (shown as an example of \ref pf_storage_entity_storage_type "\<storage_type\>" above) and attached
852 to a host called "bob" (the definition of this host is omitted here).
854 The second storage is called "Disk2", is still of the same type as Disk1 but
855 now specifies a new content file (so the contents will be different from Disk1)
856 and the filesystem uses the windows style; finally, it is attached to a second host,
857 called alice (which is again not defined here).
859 \anchor pf_storage_entity_mount
860 #### <mount> ####
863 | Attribute name | Mandatory | Values | Description |
864 | ---------------- | ----------- | -------- | ------------- |
865 | id | yes | string | Refers to a \ref pf_storage_entity_storage "<storage>" entity that will be mounted on that computer |
866 | name | yes | string | Path/location to/of the logical reference (mount point) of this disk
868 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)
869 is; this location is specified by the ``name`` attribute.
871 Here is a simple example, taken from the file ``examples/platform/storage.xml``:
874 <storage_type id="single_SSD" model="linear_no_lat" size="500GiB">
875 <model_prop id="Bwrite" value="60MBps" />
876 <model_prop id="Bread" value="200MBps" />
877 <model_prop id="Bconnection" value="220MBps" />
880 <storage id="Disk2" typeId="single_SSD"
881 content="content/win_storage_content.txt"
882 content_type="txt_windows" attach="alice" />
883 <storage id="Disk4" typeId="single_SSD"
884 content="content/small_content.txt"
885 content_type="txt_unix" attach="denise"/>
887 <host id="alice" power="1Gf">
888 <mount storageId="Disk2" name="c:"/>
891 <host id="denise" power="1Gf">
892 <mount storageId="Disk2" name="c:"/>
893 <mount storageId="Disk4" name="/home"/>
897 This example is quite interesting, as the same device, called "Disk2", is mounted by
898 two hosts at the same time! Note, however, that the host called ``alice`` is actually
899 attached to this storage, as can be seen in the \ref pf_storage_entity_storage "<storage>"
900 tag. This means that ``denise`` must access this storage through the network, but SimGrid automatically takes
901 care of that for you.
903 Furthermore, this example shows that ``denise`` has mounted two storages with different
904 filesystem types (unix and windows). In general, a host can mount as many storage devices as
908 Again, the difference between ``attach`` and ``mount`` is simply that
909 an attached storage is always physically inside (or connected to) that machine;
910 for instance, a USB stick is attached to one and only one machine (where it's plugged-in)
911 but it can only be mounted on others, as mounted storage can also be a remote location.
913 ###### Example files #####
915 \verbinclude example_filelist_xmltag_mount
917 \subsubsection pf_storage_example_files Example files
919 Several examples were already discussed above; if you're interested in full examples,
920 check the the following platforms:
922 1. ``examples/platforms/storage.xml``
923 2. ``examples/platforms/remote_io.xml``
925 If you're looking for some examplary C code, you may find the source code
926 available in the directory ``examples/msg/io/`` useful.
928 \subsubsection pf_storage_examples_modelling Modelling different situations
930 The storage functionality of SimGrid is type-agnostic, that is, the implementation
931 does not presume any type of storage, such as HDDs/SSDs, RAM,
932 CD/DVD devices, USB sticks etc.
934 This allows the user to apply the simulator for a wide variety of scenarios; one
935 common scenario would be the access of remote RAM.
937 #### Modelling the access of remote RAM ####
939 How can this be achieved in SimGrid? Let's assume we have a setup where three hosts
940 (HostA, HostB, HostC) need to access remote RAM:
950 An easy way to model this scenario is to setup and define the RAM via the
951 \ref pf_storage_entity_storage "storage" and \ref pf_storage_entity_storage_type "storage type"
952 entities and attach it to a remote dummy host; then, every host can have their own links
953 to this host (modelling for instance certain scenarios, such as PCIe ...)
958 RAM - Dummy -- Host B
963 Now, if read from this storage, the host that mounts this storage
964 communicates to the dummy host which reads from RAM and
965 sends the information back.
968 \section pf_routing Routing
970 To achieve high performance, the routing tables used within SimGrid are
971 static. This means that routing between two nodes is calculated once
972 and will not change during execution. The SimGrid team chose to use this
973 approach as it is rare to have a real deficiency of a resource;
974 most of the time, a communication fails because the links experience too much
975 congestion and hence, your connection stops before the timeout or
976 because the computer designated to be the destination of that message
979 We also chose to use shortest paths algorithms in order to emulate
980 routing. Doing so is consistent with the reality: [RIP](https://en.wikipedia.org/wiki/Routing_Information_Protocol),
981 [OSPF](https://en.wikipedia.org/wiki/Open_Shortest_Path_First), [BGP](https://en.wikipedia.org/wiki/Border_Gateway_Protocol)
982 are all calculating shortest paths. They do require some time to converge, but
983 eventually, when the routing tables have stabilized, your packets will follow
986 \subsection pf_rm Routing models
988 For each AS, you must define explicitly which routing model will
989 be used. There are 3 different categories for routing models:
991 1. \ref pf_routing_model_shortest_path "Shortest-path" based models: SimGrid calculates shortest
992 paths and manages them. Behaves more or less like most real life
994 2. \ref pf_routing_model_manual "Manually-entered" route models: you have to define all routes
995 manually in the platform description file; this can become
996 tedious very quickly, as it is very verbose.
997 Consistent with some manually managed real life routing.
998 3. \ref pf_routing_model_simple "Simple/fast models": those models offer fast, low memory routing
999 algorithms. You should consider to use this type of model if
1000 you can make some assumptions about your AS.
1001 Routing in this case is more or less ignored.
1003 \subsubsection pf_raf The router affair
1005 Using routers becomes mandatory when using shortest-path based
1006 models or when using the bindings to the ns-3 packet-level
1007 simulator instead of the native analytical network model implemented
1010 For graph-based shortest path algorithms, routers are mandatory, because these
1011 algorithms require a graph as input and so we need to have source and
1012 destination for each edge.
1014 Routers are naturally an important concept ns-3 since the
1015 way routers run the packet routing algorithms is actually simulated.
1016 SimGrid's analytical models however simply aggregate the routing time
1017 with the transfer time.
1019 So why did we incorporate routers in SimGrid? Rebuilding a graph representation
1020 only from the route information turns out to be a very difficult task, because
1021 of the missing information about how routes intersect. That is why we
1022 introduced routers, which are simply used to express these intersection points.
1023 It is important to understand that routers are only used to provide topological
1026 To express this topological information, a <b>route</b> has to be
1027 defined in order to declare which link is connected to a router.
1030 \subsubsection pf_routing_model_shortest_path Shortest-path based models
1032 The following table shows all the models that compute routes using
1033 shortest-paths algorithms are currently available in SimGrid. More detail on how
1034 to choose the best routing model is given in the Section called \"\ref pf_routing_howto_choose_wisely\".
1036 | Name | Description |
1037 | --------------------------------------------------- | -------------------------------------------------------------------------- |
1038 | \ref pf_routing_model_floyd "Floyd" | Floyd routing data. Pre-calculates all routes once |
1039 | \ref pf_routing_model_dijkstra "Dijkstra" | Dijkstra routing data. Calculates routes only when needed |
1040 | \ref pf_routing_model_dijkstracache "DijkstraCache" | Dijkstra routing data. Handles some cache for already calculated routes. |
1042 All those shortest-path models are instanciated in the same way and are
1043 completely interchangeable. Here are some examples:
1045 \anchor pf_routing_model_floyd
1050 <AS id="AS0" routing="Floyd">
1052 <cluster id="my_cluster_1" prefix="c-" suffix=""
1053 radical="0-1" power="1000000000" bw="125000000" lat="5E-5"
1054 router_id="router1"/>
1056 <AS id="AS1" routing="None">
1057 <host id="host1" power="1000000000"/>
1060 <link id="link1" bandwidth="100000" latency="0.01"/>
1062 <ASroute src="my_cluster_1" dst="AS1"
1065 <link_ctn id="link1"/>
1071 ASroute given at the end gives a topological information: link1 is
1072 between router1 and host1.
1074 #### Example platform files ####
1076 This is an automatically generated list of example files that use the Floyd
1077 routing model (the path is given relative to SimGrid's source directory)
1079 \verbinclude example_filelist_routing_floyd
1081 \anchor pf_routing_model_dijkstra
1084 #### Example platform files ####
1086 This is an automatically generated list of example files that use the Dijkstra
1087 routing model (the path is given relative to SimGrid's source directory)
1089 \verbinclude example_filelist_routing_dijkstra
1093 <AS id="AS_2" routing="Dijkstra">
1094 <host id="AS_2_host1" power="1000000000"/>
1095 <host id="AS_2_host2" power="1000000000"/>
1096 <host id="AS_2_host3" power="1000000000"/>
1097 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1098 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1099 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1100 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1101 <router id="central_router"/>
1102 <router id="AS_2_gateway"/>
1103 <!-- routes providing topological information -->
1104 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1105 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1106 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1107 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1111 \anchor pf_routing_model_dijkstracache
1112 ### DijkstraCache ###
1114 DijkstraCache example:
1116 <AS id="AS_2" routing="DijkstraCache">
1117 <host id="AS_2_host1" power="1000000000"/>
1119 (platform unchanged compared to upper example)
1122 #### Example platform files ####
1124 This is an automatically generated list of example files that use the DijkstraCache
1125 routing model (the path is given relative to SimGrid's source directory):
1127 Editor's note: At the time of writing, no platform file used this routing model - so
1128 if there are no example files listed here, this is likely to be correct.
1130 \verbinclude example_filelist_routing_dijkstra_cache
1132 \subsubsection pf_routing_model_manual Manually-entered route models
1134 | Name | Description |
1135 | ---------------------------------- | ------------------------------------------------------------------------------ |
1136 | \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. |
1138 \anchor pf_routing_model_full
1143 <AS id="AS0" routing="Full">
1144 <host id="host1" power="1000000000"/>
1145 <host id="host2" power="1000000000"/>
1146 <link id="link1" bandwidth="125000000" latency="0.000100"/>
1147 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
1151 #### Example platform files ####
1153 This is an automatically generated list of example files that use the Full
1154 routing model (the path is given relative to SimGrid's source directory):
1156 \verbinclude example_filelist_routing_full
1158 \subsubsection pf_routing_model_simple Simple/fast models
1160 | Name | Description |
1161 | ---------------------------------------- | ------------------------------------------------------------------------------ |
1162 | \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. |
1163 | \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. |
1164 | \ref pf_routing_model_vivaldi "Vivaldi" | Perfect when you want to use coordinates. Also see the corresponding \ref pf_P2P_tags "P2P section" below. |
1166 \anchor pf_routing_model_cluster
1170 In this mode, the \ref pf_cabinet "<cabinet/>" tag is available.
1172 #### Example platform files ####
1174 This is an automatically generated list of example files that use the Cluster
1175 routing model (the path is given relative to SimGrid's source directory):
1177 \verbinclude example_filelist_routing_cluster
1179 \anchor pf_routing_model_none
1182 This model does exactly what it's name advertises: Nothing. There is no routing
1183 available within this model and if you try to communicate within the AS that
1184 uses this model, SimGrid will fail unless you have explicitly activated the
1185 \ref options_model_select_network_constant "Constant Network Model" (this model charges
1186 the same for every single communication). It should
1187 be noted, however, that you can still attach an \ref pf_routing_tag_asroute "ASroute",
1188 as is demonstrated in the example below:
1190 \verbinclude platforms/cluster_and_one_host.xml
1192 #### Example platform files ####
1194 This is an automatically generated list of example files that use the None
1195 routing model (the path is given relative to SimGrid's source directory):
1197 \verbinclude example_filelist_routing_none
1200 \anchor pf_routing_model_vivaldi
1203 For more information on how to use the [Vivaldi Coordinates](https://en.wikipedia.org/wiki/Vivaldi_coordinates),
1204 see also Section \ref pf_P2P_tags "P2P tags".
1206 For documentation on how to activate this model (as some initialization must be done
1207 in the simulator), see Section \ref options_model_network_coord "Activating Coordinate Based Routing".
1209 Note that it is possible to combine the Vivaldi routing model with other routing models;
1210 an example can be found in the file \c examples/platforms/cloud.xml. This
1211 examples models an AS using Vivaldi that contains other ASes that use different
1214 #### Example platform files ####
1216 This is an automatically generated list of example files that use the None
1217 routing model (the path is given relative to SimGrid's source directory):
1219 \verbinclude example_filelist_routing_vivaldi
1222 \subsection ps_dec Defining routes
1224 There are currently four different ways to define routes:
1226 | Name | Description |
1227 | ------------------------------------------------- | ----------------------------------------------------------------------------------- |
1228 | \ref pf_routing_tag_route "route" | Used to define route between host/router |
1229 | \ref pf_routing_tag_asroute "ASroute" | Used to define route between different AS |
1230 | \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. |
1231 | \ref pf_routing_tag_bypassasroute "bypassASroute" | Used in the same way as bypassRoute, but for AS |
1233 Basically all those tags will contain an (ordered) list of references
1234 to link that compose the route you want to define.
1236 Consider the example below:
1239 <route src="Alice" dst="Bob">
1240 <link_ctn id="link1"/>
1241 <link_ctn id="link2"/>
1242 <link_ctn id="link3"/>
1246 The route here from host Alice to Bob will be first link1, then link2,
1247 and finally link3. What about the reverse route? \ref pf_routing_tag_route "Route" and
1248 \ref pf_routing_tag_asroute "ASroute" have an optional attribute \c symmetrical, that can
1249 be either \c YES or \c NO. \c YES means that the reverse route is the same
1250 route in the inverse order, and is set to \c YES by default. Note that
1251 this is not the case for bypass*Route, as it is more probable that you
1252 want to bypass only one default route.
1254 For an \ref pf_routing_tag_asroute "ASroute", things are just slightly more complicated, as you have
1255 to give the id of the gateway which is inside the AS you want to access ...
1256 So it looks like this:
1259 <ASroute src="AS1" dst="AS2"
1260 gw_src="router1" gw_dst="router2">
1261 <link_ctn id="link1"/>
1265 gw == gateway, so when any message are trying to go from AS1 to AS2,
1266 it means that it must pass through router1 to get out of the AS, then
1267 pass through link1, and get into AS2 by being received by router2.
1268 router1 must belong to AS1 and router2 must belong to AS2.
1270 \subsubsection pf_linkctn <link_ctn/>
1272 This entity has only one purpose: Refer to an already existing
1273 \ref pf_link "<link/>" when defining a route, i.e., it
1274 can only occur as a child of \ref pf_routing_tag_route "<route/>"
1276 | Attribute name | Mandatory | Values | Description |
1277 | --------------- | --------- | ------ | ----------- |
1278 | id | yes | String | The identifier of the link that should be added to the route. |
1279 | 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.
1281 #### Example Files ####
1283 This is an automatically generated list of example files that use the \c <link_ctn/>
1284 entity (the path is given relative to SimGrid's source directory):
1286 \verbinclude example_filelist_xmltag_linkctn
1288 \subsubsection pf_routing_tag_asroute ASroute
1290 The purpose of this entity is to define a route between two ASes.
1291 This is mainly useful when you're in the \ref pf_routing_model_full "Full routing model".
1293 #### Attributes ####
1295 | Attribute name | Mandatory | Values | Description |
1296 | --------------- | --------- | ------ | ----------- |
1297 | src | yes | String | The identifier of the source AS |
1298 | dst | yes | String | See the \c src attribute |
1299 | 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. |
1300 | gw_dst | yes | String | Same as \c gw_src, but with the dst AS instead. |
1301 | symmetrical | no | YES\|NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1306 <AS id="AS0" routing="Full">
1307 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
1308 radical="0-149" power="1000000000" bw="125000000" lat="5E-5"
1309 bb_bw="2250000000" bb_lat="5E-4"/>
1311 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
1312 radical="150-299" power="1000000000" bw="125000000" lat="5E-5"
1313 bb_bw="2250000000" bb_lat="5E-4"/>
1315 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1317 <ASroute src="my_cluster_1" dst="my_cluster_2"
1318 gw_src="c-my_cluster_1_router.me"
1319 gw_dst="c-my_cluster_2_router.me">
1320 <link_ctn id="backbone"/>
1322 <ASroute src="my_cluster_2" dst="my_cluster_1"
1323 gw_src="c-my_cluster_2_router.me"
1324 gw_dst="c-my_cluster_1_router.me">
1325 <link_ctn id="backbone"/>
1330 \subsubsection pf_routing_tag_route route
1332 The principle is the same as for
1333 \ref pf_routing_tag_asroute "ASroute": The route contains a list of links that
1334 provide a path from \c src to \c dst. Here, \c src and \c dst can both be either a
1335 \ref pf_host "host" or \ref pf_router "router". This is mostly useful for the
1336 \ref pf_routing_model_full "Full routing model" as well as for the
1337 \ref pf_routing_model_shortest_path "shortest-paths" based models (as they require
1338 topological information).
1341 | Attribute name | Mandatory | Values | Description |
1342 | --------------- | --------- | ---------------------- | ----------- |
1343 | src | yes | String | The value given to the source's "id" attribute |
1344 | dst | yes | String | The value given to the destination's "id" attribute. |
1345 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1350 A route in the \ref pf_routing_model_full "Full routing model" could look like this:
1352 <route src="Tremblay" dst="Bourassa">
1353 <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"/>
1357 A route in the \ref pf_routing_model_shortest_path "Shortest-Path routing model" could look like this:
1359 <route src="Tremblay" dst="Bourassa">
1364 You must only have one link in your routes when you're using them to provide
1365 topological information, as the routes here are simply the edges of the
1366 (network-)graph and the employed algorithms need to know which edge connects
1367 which pair of entities.
1369 \subsubsection pf_routing_tag_bypassasroute bypassASroute
1371 %As said before, once you choose
1372 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1373 define some of your routes, which will be specific. You may also want
1374 to bypass some routes defined in lower level AS at an upper stage:
1375 <b>bypassASroute</b> is the tag you're looking for. It allows to
1376 bypass routes defined between already defined between AS (if you want
1377 to bypass route for a specific host, you should just use byPassRoute).
1378 The principle is the same as ASroute : <b>bypassASroute</b> contains
1379 list of links that are in the path between src and dst.
1381 #### Attributes ####
1383 | Attribute name | Mandatory | Values | Description |
1384 | --------------- | --------- | ---------------------- | ----------- |
1385 | src | yes | String | The value given to the source AS's "id" attribute |
1386 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1387 | 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 |
1388 | 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|
1389 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1394 <bypassASRoute src="my_cluster_1" dst="my_cluster_2"
1395 gw_src="my_cluster_1_router"
1396 gw_dst="my_cluster_2_router">
1397 <link_ctn id="link_tmp"/>
1401 This example shows that link \c link_tmp (definition not displayed here) directly
1402 connects the router \c my_cluster_1_router in the source cluster to the router
1403 \c my_cluster_2_router in the destination router. Additionally, as the \c symmetrical
1404 attribute was not given, this route is presumed to be symmetrical.
1406 \subsubsection pf_routing_tag_bypassroute bypassRoute
1408 %As said before, once you choose
1409 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1410 define some of your routes, which will be specific. You may also want
1411 to bypass some routes defined in lower level AS at an upper stage :
1412 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
1413 routes defined between <b>host/router</b>. The principle is the same
1414 as route : <b>bypassRoute</b> contains list of links references of
1415 links that are in the path between src and dst.
1417 #### Attributes ####
1419 | Attribute name | Mandatory | Values | Description |
1420 | --------------- | --------- | ---------------------- | ----------- |
1421 | src | yes | String | The value given to the source AS's "id" attribute |
1422 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1423 | symmetrical | no | YES \| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1428 <bypassRoute src="host_1" dst="host_2">
1429 <link_ctn id="link_tmp"/>
1433 This example shows that link \c link_tmp (definition not displayed here) directly
1434 connects host \c host_1 to host \c host_2. Additionally, as the \c symmetrical
1435 attribute was not given, this route is presumed to be symmetrical.
1437 \subsection pb_baroex Basic Routing Example
1439 Let's say you have an AS named AS_Big that contains two other AS, AS_1
1440 and AS_2. If you want to make a host (h1) from AS_1 with another one
1441 (h2) from AS_2 then you'll have to proceed as follows:
1442 \li First, you have to ensure that a route is defined from h1 to the
1443 AS_1's exit gateway and from h2 to AS_2's exit gateway.
1444 \li Then, you'll have to define a route between AS_1 to AS_2. %As those
1445 AS are both resources belonging to AS_Big, then it has to be done
1446 at AS_big level. To define such a route, you have to give the
1447 source AS (AS_1), the destination AS (AS_2), and their respective
1448 gateway (as the route is effectively defined between those two
1449 entry/exit points). Elements of this route can only be elements
1450 belonging to AS_Big, so links and routers in this route should be
1451 defined inside AS_Big. If you choose some shortest-path model,
1452 this route will be computed automatically.
1454 %As said before, there are mainly 2 tags for routing :
1455 \li <b>ASroute</b>: to define routes between two <b>AS</b>
1456 \li <b>route</b>: to define routes between two <b>host/router</b>
1458 %As we are dealing with routes between AS, it means that those we'll
1459 have some definition at AS_Big level. Let consider AS_1 contains 1
1460 host, 1 link and one router and AS_2 3 hosts, 4 links and one router.
1461 There will be a central router, and a cross-like topology. At the end
1462 of the crosses arms, you'll find the 3 hosts and the router that will
1463 act as a gateway. We have to define routes inside those two AS. Let
1464 say that AS_1 contains full routes, and AS_2 contains some Floyd
1465 routing (as we don't want to bother with defining all routes). %As
1466 we're using some shortest path algorithms to route into AS_2, we'll
1467 then have to define some <b>route</b> to gives some topological
1468 information to SimGrid. Here is a file doing it all :
1471 <AS id="AS_Big" routing="Dijkstra">
1472 <AS id="AS_1" routing="Full">
1473 <host id="AS_1_host1" power="1000000000"/>
1474 <link id="AS_1_link" bandwidth="1250000000" latency="5E-4"/>
1475 <router id="AS_1_gateway"/>
1476 <route src="AS_1_host1" dst="AS_1_gateway">
1477 <link_ctn id="AS_1_link"/>
1480 <AS id="AS_2" routing="Floyd">
1481 <host id="AS_2_host1" power="1000000000"/>
1482 <host id="AS_2_host2" power="1000000000"/>
1483 <host id="AS_2_host3" power="1000000000"/>
1484 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1485 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1486 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1487 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1488 <router id="central_router"/>
1489 <router id="AS_2_gateway"/>
1490 <!-- routes providing topological information -->
1491 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1492 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1493 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1494 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1496 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1498 <ASroute src="AS_1" dst="AS_2"
1499 gw_src="AS_1_gateway"
1500 gw_dst="AS_2_gateway">
1501 <link_ctn id="backbone"/>
1506 \section pf_other_tags Tags not (directly) describing the platform
1508 The following tags can be used inside a \<platform\> tag even if they are not
1509 directly describing the platform:
1510 \li \ref pf_config "config": it allows you to pass some configuration stuff like, for
1511 example, the network model and so on. It follows the
1512 \li \ref pf_include "include": allows you to include another file into the current one.
1514 \subsection pf_config config
1516 The only purpose of this tag is to contain the \c prop tags, as described below.
1517 These tags will then configure the options as described by Section \ref options.
1520 #### Attributes ####
1522 | Attribute name | Mandatory | Values | Description |
1523 | --------------- | --------- | ---------------------- | ----------- |
1524 | id | yes | String | The identifier of the config tag when referring to id; this is basically useless, though. |
1526 #### Possible children ####
1528 Tag name | Description | Documentation
1529 ------------ | ----------- | -------------
1530 \<prop/\> | The prop tag allows you to define different configuration options following the attribute/value schema. See the \ref options page. | N/A
1535 <?xml version='1.0'?>
1536 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1537 <platform version="4">
1538 <config id="General">
1539 <prop id="maxmin/precision" value="0.000010"></prop>
1540 <prop id="cpu/optim" value="TI"></prop>
1541 <prop id="host/model" value="compound"></prop>
1542 <prop id="network/model" value="SMPI"></prop>
1543 <prop id="path" value="~/"></prop>
1544 <prop id="smpi/bw-factor" value="65472:0.940694;15424:0.697866;9376:0.58729"></prop>
1547 <AS id="AS0" routing="Full">
1551 \subsection pf_include include
1553 Even if it can be used in other contexts, this tag was originally created
1554 to be used with \ref pf_trace. The idea was to have a file describing the
1555 platform, and another file attaching traces of a given period to the platform.
1557 The drawback is that the file chuncks that will be included do not
1558 constitute valid XML files. This may explain why this feature was never really
1559 used in practice (as far as we know). Other mechanisms, such as the ability to load
1560 several platform files one after the other, could be considered in the future.
1562 In the meanwhile, the \c include tag allows you to import other platforms into your
1563 local file. This is done with the intention to help people
1564 combine their different AS and provide new platforms. Those files
1565 should contain XML that consists of
1566 \ref pf_include "include", \ref pf_cluster "cluster", \ref pf_peer "peer", \ref pf_As "AS", \ref pf_trace "trace", \ref pf_trace "tags".
1568 Do not forget to close the tag to make it work, or you will end up with an invalid XML file.
1570 #### Attributes ####
1572 | Attribute name | Mandatory | Values | Description |
1573 | --------------- | --------- | ---------------------- | ----------- |
1574 | file | yes | String | Filename of the path you want to include with either relative or absolute path. |
1579 The following example includes two files, clusterA.xml and clusterB.xml and
1580 combines them two one platform file; all hosts, routers etc. defined in
1581 each of them will then be usable.
1584 <?xml version='1.0'?>
1585 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd">
1586 <platform version="4">
1587 <AS id="main" routing="Full">
1588 <include file="clusterA.xml"></include>
1589 <include file="clusterB.xml"></include>
1594 \subsection pf_trace trace and trace_connect
1596 Both tags are an alternate way to pass files containing information on
1597 availability, state etc. to an entity. (See also, for instance, Section \ref
1598 pf_host_churn "Churn", as described for the host entity.) Instead of referring
1599 to the file directly in the host, link, or cluster tag, you proceed by defining
1600 a trace with an id corresponding to a file, later a host/link/cluster, and
1601 finally using trace_connect you say that the file trace must be used by the
1608 <AS id="AS0" routing="Full">
1609 <host id="bob" power="1000000000"/>
1611 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1612 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1616 The order here is important. \c trace_connect must come
1617 after the elements \c trace and \c host, as both the host
1618 and the trace definition must be known when \c trace_connect
1619 is parsed; the order of \c trace and \c host is arbitrary.
1622 #### \c trace attributes ####
1625 | Attribute name | Mandatory | Values | Description |
1626 | --------------- | --------- | ---------------------- | ----------- |
1627 | id | yes | String | Identifier of this trace; this is the name you pass on to \c trace_connect. |
1628 | 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. |
1629 | trace_periodicity | yes | String | This is the same as for \ref pf_host "hosts" (see there for details) |
1631 Here is an example of trace when no file name is provided:
1634 <trace id="myTrace" periodicity="1.0">
1641 #### \c trace_connect attributes ####
1643 | Attribute name | Mandatory | Values | Description |
1644 | --------------- | --------- | ---------------------- | ----------- |
1645 | kind | no | HOST_AVAIL\|POWER\|<br/>LINK_AVAIL\|BANDWIDTH\|LATENCY (Default: HOST_AVAIL) | Describes the kind of trace. |
1646 | trace | yes | String | Identifier of the referenced trace (specified of the trace's \c id attribute) |
1647 | element | yes | String | The identifier of the referenced entity as given by its \c id attribute |
1649 \section pf_hints Hints, tips and frequently requested features
1651 Now you should know at least the syntax and be able to create a
1652 platform by your own. However, after having ourselves wrote some platforms, there
1653 are some best practices you should pay attention to in order to
1654 produce good platform and some choices you can make in order to have
1655 faster simulations. Here's some hints and tips, then.
1657 @subsection Finding the platform example that you need
1659 Most platform files that we ship are in the @c examples/platforms
1660 folder. The good old @c grep tool can find the examples you need when
1661 wondering on a specific XML tag. Here is an example session searching
1662 for @ref pf_trace "trace_connect":
1665 % cd examples/platforms
1666 % grep -R -i -n --include="*.xml" "trace_connect" .
1667 ./two_hosts_platform_with_availability_included.xml:26:<trace_connect kind="SPEED" trace="A" element="Cpu A"/>
1668 ./two_hosts_platform_with_availability_included.xml:27:<trace_connect kind="HOST_AVAIL" trace="A_failure" element="Cpu A"/>
1669 ./two_hosts_platform_with_availability_included.xml:28:<trace_connect kind="SPEED" trace="B" element="Cpu B"/>
1670 ./two_hosts.xml:17: <trace_connect trace="Tremblay_power" element="Tremblay" kind="SPEED"/>
1673 \subsection pf_as_h AS Hierarchy
1674 The AS design allows SimGrid to go fast, because computing route is
1675 done only for the set of resources defined in this AS. If you're using
1676 only a big AS containing all resource with no AS into it and you're
1677 using Full model, then ... you'll loose all interest into it. On the
1678 other hand, designing a binary tree of AS with, at the lower level,
1679 only one host, then you'll also loose all the good AS hierarchy can
1680 give you. Remind you should always be "reasonable" in your platform
1681 definition when choosing the hierarchy. A good choice if you try to
1682 describe a real life platform is to follow the AS described in
1683 reality, since this kind of trade-off works well for real life
1686 \subsection pf_exit_as Exit AS: why and how
1687 Users that have looked at some of our platforms may have notice a
1688 non-intuitive schema ... Something like that :
1692 <AS id="AS_4" routing="Full">
1693 <AS id="exitAS_4" routing="Full">
1694 <router id="router_4"/>
1696 <cluster id="cl_4_1" prefix="c_4_1-" suffix="" radical="1-20" power="1000000000" bw="125000000" lat="5E-5" bb_bw="2250000000" bb_lat="5E-4"/>
1697 <cluster id="cl_4_2" prefix="c_4_2-" suffix="" radical="1-20" power="1000000000" bw="125000000" lat="5E-5" bb_bw="2250000000" bb_lat="5E-4"/>
1698 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1699 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1700 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1701 <ASroute src="cl_4_1"
1703 gw_src="c_4_1-cl_4_1_router"
1704 gw_dst="c_4_2-cl_4_2_router">
1705 <link_ctn id="4_1"/>
1706 <link_ctn id="bb_4"/>
1707 <link_ctn id="4_2"/>
1709 <ASroute src="cl_4_1"
1711 gw_src="c_4_1-cl_4_1_router"
1713 <link_ctn id="4_1"/>
1714 <link_ctn id="bb_4"/>
1716 <ASroute src="cl_4_2"
1718 gw_src="c_4_2-cl_4_2_router"
1720 <link_ctn id="4_2"/>
1721 <link_ctn id="bb_4"/>
1726 In the AS_4, you have an exitAS_4 defined, containing only one router,
1727 and routes defined to that AS from all other AS (as cluster is only a
1728 shortcut for an AS, see cluster description for details). If there was
1729 an upper AS, it would define routes to and from AS_4 with the gateway
1730 router_4. It's just because, as we did not allowed (for performances
1731 issues) to have routes from an AS to a single host/router, you have to
1732 enclose your gateway, when you have AS included in your AS, within an
1733 AS to define routes to it.
1735 \subsection pf_P2P_tags P2P or how to use coordinates
1736 SimGrid allows you to use some coordinated-based system, like vivaldi,
1737 to describe a platform. The main concept is that you have some peers
1738 that are located somewhere: this is the function of the
1739 <b>coordinates</b> of the \<peer\> or \<host\> tag. There's nothing
1740 complicated in using it, here is an example of it:
1743 <?xml version='1.0'?>
1744 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1745 <platform version="4">
1747 <config id="General">
1748 <prop id="network/coordinates" value="yes"></prop>
1750 <AS id="AS0" routing="Vivaldi">
1751 <host id="100030591" coordinates="25.5 9.4 1.4" power="1.5Gf" />
1752 <host id="100036570" coordinates="-12.7 -9.9 2.1" power="7.3Gf" />
1754 <host id="100429957" coordinates="17.5 6.7 18.8" power="8.3Gf" />
1759 Coordinates are then used to calculate latency (in microseconds)
1760 between two hosts by calculating the distance between the two hosts
1761 coordinates with the following formula: distance( (x1, y1, z1), (x2,
1762 y2, z2) ) = euclidian( (x1,y1), (x2,y2) ) + abs(z1) + abs(z2)
1764 In other words, we take the euclidian distance on the two first
1765 dimensions, and then add the absolute values found on the third
1766 dimension. This may seem strange, but it was found to allow better
1767 approximations of the latency matrices (see the paper describing
1770 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between AS.
1771 That is for example what is commonly done when using peers (see Section \ref pf_peer).
1773 <?xml version='1.0'?>
1774 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1775 <platform version="4">
1777 <config id="General">
1778 <prop id="network/coordinates" value="yes"></prop>
1780 <AS id="AS0" routing="Vivaldi">
1781 <peer id="peer-0" coordinates="173.0 96.8 0.1" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1782 <peer id="peer-1" coordinates="247.0 57.3 0.6" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1783 <peer id="peer-2" coordinates="243.4 58.8 1.4" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1787 In such a case though, we connect the AS created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1788 This means that to route between AS1 and AS2, it will use the coordinates of router_AS1 and router_AS2.
1789 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1790 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.
1793 \subsection pf_routing_howto_choose_wisely Choosing wisely the routing model to use
1796 Choosing wisely the routing model to use can significantly fasten your
1797 simulation/save your time when writing the platform/save tremendous
1798 disk space. Here is the list of available model and their
1799 characteristics (lookup : time to resolve a route):
1801 \li <b>Full</b>: Full routing data (fast, large memory requirements,
1803 \li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1804 lookup, lesser memory requirements, shortest path routing only).
1805 Calculates all routes at once at the beginning.
1806 \li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1807 lookup, small memory requirements, shortest path routing only).
1808 Calculates a route when necessary.
1809 \li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1810 fast lookup, small memory requirements, shortest path routing
1811 only). Same as Dijkstra, except it handles a cache for latest used
1813 \li <b>None</b>: No routing (usable with Constant network only).
1814 Defines that there is no routes, so if you try to determine a
1815 route without constant network within this AS, SimGrid will raise
1817 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1818 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1821 \subsection pf_switch I want to describe a switch but there is no switch tag!
1823 Actually we did not include switch tag. But when you're trying to
1824 simulate a switch, assuming
1825 fluid bandwidth models are used (which SimGrid uses by default unless
1826 ns-3 or constant network models are activated), the limiting factor is
1827 switch backplane bandwidth. So, essentially, at least from
1828 the simulation perspective, a switch is similar to a
1829 link: some device that is traversed by flows and with some latency and
1830 so,e maximum bandwidth. Thus, you can simply simulate a switch as a
1832 can be connected to this "switch", which is then included in routes just
1836 \subsection pf_multicabinets I want to describe multi-cabinets clusters!
1838 You have several possibilities, as usual when modeling things. If your
1839 cabinets are homogeneous and the intercabinet network negligible for
1840 your study, you should just create a larger cluster with all hosts at
1843 In the rare case where your hosts are not homogeneous between the
1844 cabinets, you can create your cluster completely manually. For that,
1845 create an As using the Cluster routing, and then use one
1846 <cabinet> for each cabinet. This cabinet tag can only be used an
1847 As using the Cluster routing schema, and creating
1849 Be warned that creating a cluster manually from the XML with
1850 <cabinet>, <backbone> and friends is rather tedious. The
1851 easiest way to retrieve some control of your model without diving into
1852 the <cluster> internals is certainly to create one separate
1853 <cluster> per cabinet and interconnect them together. This is
1854 what we did in the G5K example platform for the Graphen cluster.
1856 \subsection pf_platform_multipath I want to express multipath routing in platform files!
1858 It is unfortunately impossible to express the fact that there is more
1859 than one routing path between two given hosts. Let's consider the
1860 following platform file:
1863 <route src="A" dst="B">
1866 <route src="B" dst="C">
1869 <route src="A" dst="C">
1874 Although it is perfectly valid, it does not mean that data traveling
1875 from A to C can either go directly (using link 3) or through B (using
1876 links 1 and 2). It simply means that the routing on the graph is not
1877 trivial, and that data do not following the shortest path in number of
1878 hops on this graph. Another way to say it is that there is no implicit
1879 in these routing descriptions. The system will only use the routes you
1880 declare (such as <route src="A" dst="C"><link_ctn
1881 id="3"/></route>), without trying to build new routes by aggregating
1884 You are also free to declare platform where the routing is not
1885 symmetrical. For example, add the following to the previous file:
1888 <route src="C" dst="A">
1894 This makes sure that data from C to A go through B where data from A
1895 to C go directly. Don't worry about realism of such settings since
1896 we've seen ways more weird situation in real settings (in fact, that's
1897 the realism of very regular platforms which is questionable, but
1898 that's another story).
