1 /*! \page platform %Model the underlying 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
7 want to simulate your application and lastly information about the deployment
8 process: Which process should be deployed to which processor/core?
10 For the last two items, there are essentially two possible ways you can provide
11 this information as an input:
12 \li You can program, if you're using MSG, some of MSG's platform and
13 deployment functions (\ref msg_simulation). If you want to use this,
14 check the particular documentation. (You can also check the section
15 \ref pf_flexml_bypassing, however, this documentation is deprecated;
16 there is a new, but undocumented, way to do it properly).
17 \li You can use two XML files: one contains the platform description while
18 the other contains the deployment instructions. The platform description
19 can also be in Lua format.
21 For more information on SimGrid's deployment features, please refer to
22 the \ref deployment documentation.
24 The platform description may be intricate. This documentation is all
25 about how to write this file: The basic concepts are introduced. Furthermore,
26 advanced options are explained. Additionally, some hints and tips on how to
27 write a good platform description are given.
29 \section pf_overview Some words about XML and DTD
31 We chose to use XML not only because it's extensible but also because many
32 tools (and plugins for existing tools) are available that facilitate editing and
33 validating XML files. Furthermore, libraries that parse XML are often already
34 available and very well tested.
36 The XML checking is done based on the Document Type Definition (DTD) file,
38 <a href="http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd">http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd</a>.
40 If you read the DTD, you should notice the following:
41 \li The platform tags contain a version attribute; the current version is 4.
42 This property might be used in the future to provide backwards
44 \li The DTD contains definitions for the two files used by SimGrid (i.e.,
45 platform description and deployment).
47 \section pf_basics Basic concepts
49 Nowadays, the Internet is composed of a bunch of independently managed
50 networks. Within each of those networks, there are entry and exit
51 points (most of the time, you can both enter and exit through the same
52 point); this allows to leave the current network and reach other
53 networks, possibly even in other locations.
54 At the upper level, such a network is called
55 <b>Autonomous System (AS)</b>, while at the lower level it is named
56 sub-network, or LAN (local area network).
57 They are indeed autonomous: routing is defined
58 (within the limits of his network) by the administrator, and so, those
59 networks can operate without a connection to other
60 networks. So-called gateways allow you to go from one network to
61 another, if such a (physical) connection exists. Every node in one network
62 that can be directly reached (i.e., without traversing other nodes) from
63 another network is called a gateway.
64 Each autonomous system consists of equipment such as cables (network links),
65 routers and switches as well as computers.
67 The structure of the SimGrid platform description relies exactly on the same
68 concept as a real-life platform (see above). Every resource (computers,
69 network equipment etc.) belongs to an AS, which can be defined by using the
70 \<AS\> tag. Within an AS, the routing between its elements can be defined
71 abitrarily. There are several modes for routing, and exactly one mode must be
72 selected by specifying the routing attribute in the AS tag:
75 <AS id="AS0" routing="Full">
79 Other supported values for the routing attribute can be found below, Section
82 There is also the ``<route>`` tag; this tag takes two attributes, ``src`` (source)
83 and ``dst`` (destination). Both source and destination must be valid identifiers
84 for routers (these will be introduced later). Contained by the ``<route>`` are
85 network links; these links must be used in order to communicate from the source
86 to the destination specified in the tag. Hence, a route merely describes
87 how to reach a router from another router.
90 More information and (code-)examples can be found in Section \ref pf_rm.
92 An AS can also contain itself one or more AS; this allows you to
93 model the hierarchy of your platform.
95 ### Within each AS, the following types of resources exist:
97 %Resource | Documented in Section | Description
98 --------------- | --------------------- | -----------
99 AS | | Every Autonomous System (AS) may contain one or more AS.
100 host | \ref pf_host | This entity carries out the actual computation. For this reason, it contains processors (with potentially multiple cores).
101 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.
102 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.
103 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.
105 %As it is desirable to interconnect these resources, a routing has to be
106 defined. The AS is supposed to be Autonomous, hence this has to be done at the
107 AS level. The AS handles two different types of entities (<b>host/router</b>
108 and <b>AS</b>). However, the user is responsible to define routes between those resources,
109 otherwise entities will be unconnected and therefore unreachable from other
110 entities. Although several routing algorithms are built into SimGrid (see
111 \ref pf_rm), you might encounter a case where you want to define routes
112 manually (for instance, due to specific requirements of your platform).
114 There are three tags to use:
115 \li <b>ASroute</b>: to define routes between two <b>AS</b>
116 \li <b>route</b>: to define routes between two <b>host/router</b>
117 \li <b>bypassRoute</b>: to define routes between two <b>AS</b> that
118 will bypass default routing (as specified by the ``routing`` attribute
119 supplied to ``<AS>``, see above).
121 Here is an illustration of these concepts:
123 
124 Circles represent processing units and squares represent network routers. Bold
125 lines represent communication links. AS2 models the core of a national
126 network interconnecting a small flat cluster (AS4) and a larger
127 hierarchical cluster (AS5), a subset of a LAN (AS6), and a set of peers
128 scattered around the world (AS7).
130 \section pf_pftags Resource description
132 \subsection pf_As Platform: The <AS> tag
134 The concept of an AS was already outlined above (Section \ref pf_basics);
135 recall that the AS is so important because it groups other resources (such
136 as routers/hosts) together (in fact, these resources must be contained by
139 Available attributes :
141 Attribute name | Mandatory | Values | Description
142 --------------- | --------- | ------ | -----------
143 id | yes | String | The identifier of an AS; facilitates referring to this AS. ID must be unique.
144 routing | yes | Full\| Floyd\| Dijkstra\| DijkstraCache\| None\| Vivaldi\| Cluster | See Section \ref pf_rm for details.
149 <AS id="AS0" routing="Full">
150 <host id="host1" power="1000000000"/>
151 <host id="host2" power="1000000000"/>
152 <link id="link1" bandwidth="125000000" latency="0.000100"/>
153 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
157 In this example, AS0 contains two hosts (host1 and host2). The route
158 between the hosts goes through link1.
160 \subsection pf_Cr Computing resources: hosts, clusters and peers.
162 \subsubsection pf_host <host/>
164 A <b>host</b> represents a computer/node card. Every host is able to execute
165 code and it can send and receive data to/from other hosts. Most importantly,
166 a host can contain more than 1 core.
170 Attribute name | Mandatory | Values | Description
171 --------------- | --------- | ------ | -----------
172 id | yes | String | The identifier of the host. facilitates referring to this AS.
173 power | yes | double (must be > 0.0) | Computational power of every core of this host in FLOPS. Must be larger than 0.0.
174 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.
175 availability | no | int | <b>Specify if the percentage of power available.</b> (What? TODO)
176 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.
177 state | no | ON\|OFF<br/> (Default: ON) | Is this host running or not?
178 state_file | no | string | Same mechanism as availability_file.<br /> <b>Note:</b> The filename must be specified with your system's format.
179 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.
181 ### Possible children: ###
183 Tag name | Description | Documentation
184 ------------ | ----------- | -------------
185 \<mount/\> | Defines mounting points between some storage resource and the host. | \ref pf_storage_entity_mount
186 \<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
191 <host id="host1" power="1000000000"/>
192 <host id="host2" power="1000000000">
193 <prop id="color" value="blue"/>
194 <prop id="rendershape" value="square"/>
199 \anchor pf_host_dynamism
200 ### Expressing dynamism ###
202 SimGrid provides mechanisms to change a hosts' availability over
203 time, using the ``availability_file`` attribute to the ``\<host\>`` tag
204 and a separate text file whose syntax is exemplified below.
206 #### Adding a trace file ####
209 <platform version="4">
210 <host id="bob" power="500Gf" availability_file="bob.trace" />
214 #### Example of "bob.trace" file ####
223 Let us begin to explain this example by looking at line 2. (Line 1 will become clear soon).
224 The first column describes points in time, in this case, time 0. The second column
225 describes the relative amount of power this host is able to deliver (relative
226 to the maximum performance specified in the ``\<host\>`` tag). (Clearly, the
227 second column needs to contain values that are not smaller than 0 and not larger than 1).
228 In this example, our host will deliver 500 Mflop/s at time 0, as 500 Mflop/s is the
229 maximum performance of this host. At time 11.0, it will
230 deliver half of its maximum performance, i.e., 250 Mflop/s until time 20.0 when it will
231 will start delivering 80\% of its power. In this example, this amounts to 400 Mflop/s.
233 Since the periodicity in line 1 was set to be 1.0, i.e., 1 timestep, this host will
234 continue to provide 500 Mflop/s from time 21. From time 32 it will provide 250 MFlop/s and so on.
236 ### Changing initial state ###
238 It is also possible to specify whether the host is up or down by setting the
239 ``state`` attribute to either <b>ON</b> (default value) or <b>OFF</b>.
241 #### Example: Expliciting the default value "ON" ####
244 <platform version="4">
245 <host id="bob" power="500Gf" state="ON" />
249 If you want this host to be unavailable, simply substitute ON with OFF.
251 \anchor pf_host_churn
252 ### Expressing churn ###
254 To express the fact that a host can change state over time (as in P2P
255 systems, for instance), it is possible to use a file describing the time
256 at which the host is turned on or off. An example of the content
257 of such a file is presented below.
259 #### Adding a state file ####
262 <platform version="4">
263 <host id="bob" power="500Gf" state_file="bob.fail" />
267 #### Example of "bob.fail" file ####
275 A zero value means <b>down</b> (i.e., OFF) while a positive one means <b>up and
276 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
277 turned off and at time 2.0, it is turned on again until time 12 (2 plus the
278 periodicity 10). It will be turned off again at time 13.0 until time 23.0, and
282 \subsubsection pf_cluster <cluster>
284 ``<cluster />`` represents a machine-cluster. It is most commonly used
285 when one wants to define many hosts and a network quickly. Technically,
286 ``cluster`` is a meta-tag: <b>from the inner SimGrid point of
287 view, a cluster is an AS where some optimized routing is defined</b>.
288 The default inner organization of the cluster is as follow:
294 ____________|__________|_____________ backbone
296 l0| l1| l2| l97| l96 | | l99
302 Here, a set of <b>host</b>s is defined. Each of them has a <b>link</b>
303 to a central backbone (backbone is a link itself, as a link can
304 be used to represent a switch, see the switch / link section
305 below for more details about it). A <b>router</b> allows to connect a
306 <b>cluster</b> to the outside world. Internally,
307 SimGrid treats a cluster as an AS containing all hosts: the router is the default
308 gateway for the cluster.
310 There is an alternative organization, which is as follows:
324 The principle is the same, except that there is no backbone. This representation
325 can be obtained easily: just do not set the bb_* attributes.
328 Attribute name | Mandatory | Values | Description
329 --------------- | --------- | ------ | -----------
330 id | yes | string | The identifier of the cluster. Facilitates referring to this cluster.
331 prefix | yes | string | Each node of the cluster has to have a name. This name will be prefixed with this prefix.
332 suffix | yes | string | Each node of the cluster will be suffixed with this suffix
333 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.
334 power | yes | int | Same as the ``power`` attribute of the ``\<host\>`` tag.
335 core | no | int (default: 1) | Same as the ``core`` attribute of the ``\<host\>`` tag.
336 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
337 lat | yes | int | Latency for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
338 sharing_policy | no | string | Sharing policy for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
339 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>).
340 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>).
341 bb_sharing_policy | no | string | Sharing policy for the backbone (if any). See <b>link</b> section for syntax/details.
342 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>
343 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".
344 loopback_lat | no | int | Latency for loopback (if any). See <b>link</b> section for syntax/details. See loopback_bw for more info.
345 topology | no | FLAT\|TORUS\|FAT_TREE (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> and FAT_TREE attributes for this tag.
346 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".
350 the router name is defined as the resulting String in the following
354 router_name = prefix + clusterId + _router + suffix;
358 #### Cluster example ####
360 Consider the following two (and independent) uses of the ``cluster`` tag:
363 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
364 power="1e9" bw="125e6" lat="5E-5"/>
366 <cluster id="my_cluster_2" prefix="c-" suffix=".me" radical="0-99"
367 power="1e9" bw="125e6" lat="5E-5"
368 bb_bw="2.25e9" bb_lat="5E-4"/>
371 The second example creates one router and 100 machines with the following names:
373 c-my_cluster_2_router.me
381 \subsubsection pf_cabinet <cabinet>
384 This tag is only available when the routing mode of the AS
385 is set to ``Cluster``.
387 The ``<cabinet />`` tag is, like the \ref pf_cluster "<cluster>" tag,
388 a meta-tag. This means that it is simply a shortcut for creating a set of (homogenous) hosts and links quickly;
389 unsurprisingly, this tag was introduced to setup cabinets in data centers quickly. Unlike
390 <cluster>, however, the <cabinet> assumes that you create the backbone
391 and routers yourself; see our examples below.
395 Attribute name | Mandatory | Values | Description
396 --------------- | --------- | ------ | -----------
397 id | yes | string | The identifier of the cabinet. Facilitates referring to this cluster.
398 prefix | yes | string | Each node of the cabinet has to have a name. This name will be prefixed with this prefix.
399 suffix | yes | string | Each node of the cabinet will be suffixed with this suffix
400 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.
401 power | yes | int | Same as the ``power`` attribute of the \ref pf_host "<host>" tag.
402 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
403 lat | yes | int | Latency for the links between nodes and backbone (if any). See the \ref pf_link "link section" for syntax/details.
406 Please note that as of now, it is impossible to change attributes such as,
407 amount of cores (always set to 1), the initial state of hosts/links
408 (always set to ON), the sharing policy of the links (always set to \ref pf_sharing_policy_fullduplex "FULLDUPLEX").
412 The following example was taken from ``examples/platforms/meta_cluster.xml`` and
413 shows how to use the cabinet tag.
416 <AS id="my_cluster1" routing="Cluster">
417 <cabinet id="cabinet1" prefix="host-" suffix=".cluster1"
418 power="1Gf" bw="125MBps" lat="100us" radical="1-10"/>
419 <cabinet id="cabinet2" prefix="host-" suffix=".cluster1"
420 power="1Gf" bw="125MBps" lat="100us" radical="11-20"/>
421 <cabinet id="cabinet3" prefix="host-" suffix=".cluster1"
422 power="1Gf" bw="125MBps" lat="100us" radical="21-30"/>
424 <backbone id="backbone1" bandwidth="2.25GBps" latency="500us"/>
429 Please note that you must specify the \ref pf_backbone "<backbone>"
430 tag by yourself; this is not done automatically and there are no checks
431 that ensure this backbone was defined.
433 The hosts generated in the above example are named host-1.cluster, host-2.cluster1
437 \subsubsection pf_peer The <peer> tag
439 This tag represents a peer, as in Peer-to-Peer (P2P) networks. However, internally,
440 SimGrid transforms a peer into an AS (similar to Cluster). Hence, this tag
441 is virtually only a shortcut that comes with some pre-defined resources
442 and values. These are:
444 \li A tiny AS whose routing type is cluster is created
446 \li Two links: One for download and one for upload. This is
447 convenient to use and simulate stuff under the last mile model (e.g., ADSL peers).
448 \li It connects the two links to the host
449 \li It creates a router (a gateway) that serves as an entry point for this peer zone.
450 This router has coordinates.
454 Attribute name | Mandatory | Values | Description
455 --------------- | --------- | ------ | -----------
456 id | yes | string | The identifier of the peer. Facilitates referring to this peer.
457 power | yes | int | See the description of the ``host`` tag for this attribute
458 bw_in | yes | int | Bandwidth downstream
459 bw_out | yes | int | Bandwidth upstream
460 lat | yes | double | Latency for both up- and downstream, in seconds.
461 coordinates | no | string | Coordinates of the gateway for this peer. Example value: 12.8 14.4 6.4
462 sharing_policy | no | SHARED\|FULLDUPLEX (default: FULLDUPLEX) | Sharing policy for links. See <b>link</b> description for details.
463 availability_file| no | string | Availability file for the peer. Same as host availability file. See <b>host</b> description for details.
464 state_file | no | string | State file for the peer. Same as host state file. See <b>host</b> description for details.
466 Internally, SimGrid transforms any ``\<peer/\>`` construct such as
469 coordinates="12.8 14.4 6.4"
475 into an ``\<AS\>`` (see Sections \ref pf_basics and \ref pf_As). In fact, this example of the ``\<peer/\>`` tag
476 is completely equivalent to the following declaration:
479 <AS id="as_FOO" routing="Cluster">
480 <host id="peer_FOO" power="1.5Gf"/>
481 <link id="link_FOO_UP" bandwidth="2.25GBps" latency="500us"/>
482 <link id="link_FOO_DOWN" bandwidth="2.25GBps" latency="500us"/>
483 <router id="router_FOO" coordinates="25.5 9.4 1.4"/>
484 <host_link id="peer_FOO" up="link_FOO_UP" down="link_FOO_DOWN"/>
489 \subsection pf_ne Network equipments: links and routers
491 There are two tags at all times available to represent network entities and
492 several other tags that are available only in certain contexts.
493 1. ``<link>``: Represents a entity that has a limited bandwidth, a
494 latency, and that can be shared according to TCP way to share this
497 The concept of links in SimGrid may not be intuitive, as links are not
498 limited to connecting (exactly) two entities; in fact, you can have more than
499 two equipments connected to it. (In graph theoretical terms: A link in
500 SimGrid is not an edge, but a hyperedge)
502 2. ``<router/>``: Represents an entity that a message can be routed
503 to, but that is unable to execute any code. In SimGrid, routers have also
504 no impact on the performance: Routers do not limit any bandwidth nor
505 do they increase latency. As a matter of fact, routers are (almost) ignored
506 by the simulator when the simulation has begun.
508 3. ``<backbone/>``: This tag is only available when the containing AS is
509 used as a cluster (i.e., mode="Cluster")
512 If you want to represent an entity like a switch, you must use ``<link>`` (see section). Routers are used
513 to run some routing algorithm and determine routes (see Section \ref pf_routing for details).
515 \subsubsection pf_router <router/>
517 %As said before, <b>router</b> is used only to give some information
518 for routing algorithms. So, it does not have any attributes except :
522 Attribute name | Mandatory | Values | Description
523 --------------- | --------- | ------ | -----------
524 id | yes | string | The identifier of the router to be used when referring to it.
525 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.
530 <router id="gw_dc1_horizdist"/>
533 \subsubsection pf_link <link/>
535 Network links can represent one-hop network connections. They are
536 characterized by their id and their bandwidth; links can (but may not) be subject
541 Attribute name | Mandatory | Values | Description
542 --------------- | --------- | ------ | -----------
543 id | yes | string | The identifier of the link to be used when referring to it.
544 bandwidth | yes | int | Maximum bandwidth for this link, given in bytes/s
545 latency | no | double (default: 0.0) | Latency for this link.
546 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.
547 state | no | ON\|OFF (default: ON) | Allows you to to turn this link on or off (working / not working)
548 bandwidth_file | no | string | Allows you to use a file as input for bandwidth.
549 latency_file | no | string | Allows you to use a file as input for latency.
550 state_file | no | string | Allows you to use a file as input for states.
553 #### Possible shortcuts for ``latency`` ####
555 When using the latency attribute, you can specify the latency by using the scientific
556 notation or by using common abbreviations. For instance, the following three tags
560 <link id="LINK1" bandwidth="125000000" latency="5E-6"/>
561 <link id="LINK1" bandwidth="125000000" latency="5us"/>
562 <link id="LINK1" bandwidth="125000000" latency="0.000005"/>
565 Here, the second tag uses "us", meaning "microseconds". Other shortcuts are:
567 Name | Abbreviation | Time (in seconds)
568 ---- | ------------ | -----------------
569 Week | w | 7 * 24 * 60 * 60
570 Day | d | 24 * 60 * 60
574 Millisecond | ms | 0.001 = 10^(-3)
575 Microsecond | us | 0.000001 = 10^(-6)
576 Nanosecond | ns | 0.000000001 = 10^(-9)
577 Picosecond | ps | 0.000000000001 = 10^(-12)
579 #### Sharing policy ####
581 \anchor sharing_policy_shared
582 By default a network link is \b SHARED, i.e., if two or more data flows go
583 through a link, the bandwidth is shared fairly among all data flows. This
584 is similar to the sharing policy TCP uses.
586 \anchor pf_sharing_policy_fatpipe
587 On the other hand, if a link is defined as a \b FATPIPE,
588 each flow going through this link will be provided with the complete bandwidth,
589 i.e., no sharing occurs and the bandwidth is only limiting each flow individually.
590 Please note that this is really on a per-flow basis, not only on a per-host basis!
591 The complete bandwidth provided by this link in this mode
592 is ``number_of_flows*bandwidth``, with at most ``bandwidth`` being available per flow.
594 Using the FATPIPE mode allows to model backbones that won't affect performance
597 \anchor pf_sharing_policy_fullduplex
598 The last mode available is \b FULLDUPLEX. This means that SimGrid will
599 automatically generate two links (one carrying the suffix _UP and the other the
600 suffix _DOWN) for each ``<link>`` tag. This models situations when the direction
601 of traffic is important.
604 Transfers from one side to the other will interact similarly as
605 TCP when ACK returning packets circulate on the other direction. More
606 discussion about it is available in the description of link_ctn description.
608 In other words: The SHARED policy defines a physical limit for the bandwidth.
609 The FATPIPE mode defines a limit for each application,
610 with no upper total limit.
613 Tip: By using the FATPIPE mode, you can model big backbones that
614 won't affect performance (except latency).
619 <link id="SWITCH" bandwidth="125000000" latency="5E-5" sharing_policy="FATPIPE" />
622 #### Expressing dynamism and failures ####
624 Similar to hosts, it is possible to declare links whose state, bandwidth
625 or latency changes over time (see Section \ref pf_host_dynamism for details).
627 In the case of network links, the ``bandwidth`` and ``latency`` attributes are
628 replaced by the ``bandwidth_file`` and ``latency_file`` attributes.
629 The following XML snippet demonstrates how to use this feature in the platform
630 file. The structure of the files "link1.bw" and "link1.lat" is shown below.
633 <link id="LINK1" state_file="link1.fail" bandwidth="80000000" latency=".0001" bandwidth_file="link1.bw" latency_file="link1.lat" />
637 Even if the syntax is the same, the semantic of bandwidth and latency
638 trace files differs from that of host availability files. For bandwidth and
639 latency, the corresponding files do not
640 express availability as a fraction of the available capacity but directly in
641 bytes per seconds for the bandwidth and in seconds for the latency. This is
642 because most tools allowing to capture traces on real platforms (such as NWS)
643 express their results this way.
645 ##### Example of "link1.bw" file #####
653 In this example, the bandwidth changes repeatedly, with all changes
654 being repeated every 12 seconds.
656 At the beginning of the the simulation, the link's bandwidth is 80,000,000
657 B/s (i.e., 80 Mb/s); this value was defined in the XML snippet above.
658 After four seconds, it drops to 40 Mb/s (line 2), and climbs
659 back to 60 Mb/s after another 4 seconds (line 3). The value does not change any
660 more until the end of the period, that is, after 12 seconds have been simulated).
661 At this point, periodicity kicks in and this behavior is repeated: Seconds
662 12-16 will experience 80 Mb/s, 16-20 40 Mb/s etc.).
664 ##### Example of "link1.lat" file #####
673 In this example, the latency varies with a period of 5 seconds.
674 In the xml snippet above, the latency is initialized to be 0.0001s (100µs). This
675 value will be kept during the first second, since the latency_file contains
676 changes to this value at second one, two and three.
677 At second one, the value will be 0.001, i.e., 1ms. One second later it will
678 be adjusted to 0.01 (or 10ms) and one second later it will be set again to 1ms. The
679 value will not change until second 5, when the periodicity defined in line 1
680 kicks in. It then loops back, starting at 100µs (the initial value) for one second.
683 #### The ``<prop/>`` tag ####
685 Similar to the ``<host>`` tag, a link may also contain the ``<prop/>`` tag; see the host
686 documentation (Section \ref pf_host) for an example.
689 \subsubsection pf_backbone <backbone/>
692 This tag is <b>only available</b> when the containing AS uses the "Cluster" routing mode!
694 Using this tag, you can designate an already existing link to be a backbone.
696 Attribute name | Mandatory | Values | Description
697 --------------- | --------- | ------ | -----------
698 id | yes | string | Name of the link that is supposed to act as a backbone.
700 \subsection pf_storage Storage
703 This is a prototype version that should evolve quickly, hence this
704 is just some doc valuable only at the time of writing.
705 This section describes the storage management under SimGrid ; nowadays
706 it's only usable with MSG. It relies basically on linux-like concepts.
707 You also may want to have a look to its corresponding section in \ref
708 msg_file_management ; access functions are organized as a POSIX-like
711 \subsubsection pf_sto_conc Storage - Main Concepts
713 The storage facilities implemented in SimGrid help to model (and account for)
714 storage devices, such as tapes, hard-drives, CD or DVD devices etc.
715 A typical situation is depicted in the figure below:
717 \image html ./webcruft/storage_sample_scenario.png
718 \image latex ./webcruft/storage_sample_scenario.png "storage_sample_scenario" width=\textwidth
720 In this figure, two hosts called Bob and Alice are interconnected via a network
721 and each host is physically attached to a disk; it is not only possible for each host to
722 mount the disk they are attached to directly, but they can also mount disks
723 that are in a remote location. In this example, Bob mounts Alice's disk remotely
724 and accesses the storage via the network.
726 SimGrid provides 3 different entities that can be used to model setups
727 that include storage facilities:
729 Entity name | Description
730 --------------- | -----------
731 \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).
732 \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.
733 \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).
736 \anchor pf_storage_content_file
737 ### %Storage Content File ###
739 In order to assess exactly how much time is spent reading from the storage,
740 SimGrid needs to know what is stored on the storage device (identified by distinct (file-)name, like in a file system)
741 and what size this content has.
744 The content file is never changed by the simulation; it is parsed once
745 per simulation and kept in memory afterwards. When the content of the
746 storage changes, only the internal SimGrid data structures change.
748 \anchor pf_storage_content_file_structure
749 #### Structure of a %Storage Content File ####
751 Here is an excerpt from two storage content file; if you want to see the whole file, check
752 the file ``examples/platforms/content/storage_content.txt`` that comes with the
755 SimGrid essentially supports two different formats: UNIX-style filepaths should
756 follow the well known format:
759 /lib/libsimgrid.so.3.6.2 12710497
763 /bin/simgrid_update_xml 5018
764 /bin/graphicator 66986
765 /bin/simgrid-colorizer 2993
770 Windows filepaths, unsurprisingly, use the windows style:
773 \Windows\avastSS.scr 41664
774 \Windows\bfsvc.exe 75264
775 \Windows\bootstat.dat 67584
776 \Windows\CoreSingleLanguage.xml 31497
778 \Windows\dchcfg64.exe 335464
779 \Windows\dcmdev64.exe 93288
783 The different file formats come at a cost; in version 3.12 (and most likely
784 in later versions, too), copying files from windows-style storages to unix-style
785 storages (and vice versa) is not supported.
787 \anchor pf_storage_content_file_create
788 #### Generate a %Storage Content File ####
790 If you want to generate a storage content file based on your own filesystem (or at least a filesystem you have access to),
791 try running this command (works only on unix systems):
794 find . -type f -exec ls -1s --block=1 {} \; 2>/dev/null | awk '{ print $2 " " $1}' > ./content.txt
797 \subsubsection pf_storage_entities The Storage Entities
799 These are the entities that you can use in your platform files to include
800 storage in your model. See also the list of our \ref pf_storage_example_files "example files";
801 these might also help you to get started.
803 \anchor pf_storage_entity_storage_type
804 #### \<storage_type\> ####
806 Attribute name | Mandatory | Values | Description
807 --------------- | --------- | ------ | -----------
808 id | yes | string | Identifier of this storage_type; used when referring to it
809 model | yes | string | For reasons of future backwards compatibility only; specifies the name of the model for the storage that should be used
810 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)
811 content | yes | string | Path to a \ref pf_storage_content_file "Storage Content File" on your system. This file must exist.
812 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"!)
814 This tag must contain some predefined model properties, specified via the <model_prop> tag. Here is a list,
815 see below for an example:
817 Property id | Mandatory | Values | Description
818 --------------- | --------- | ------ | -----------
819 Bwrite | yes | string | Bandwidth for write access; in B/s (but you can also specify e.g. "30MBps")
820 Bread | yes | string | Bandwidth for read access; in B/s (but you can also specify e.g. "30MBps")
821 Bconnexion | yes | string | Throughput (of the storage connector) in B/s.
824 A storage_type can also contain the <b><prop></b> tag. The <prop> tag allows you
825 to associate additional information to this <storage_type> and follows the
826 attribute/value schema; see the example below. You may want to use it to give information to
827 the tool you use for rendering your simulation, for example.
829 Here is a complete example for the ``storage_type`` tag:
831 <storage_type id="single_HDD" model="linear_no_lat" size="4000" content_type="txt_unix">
832 <model_prop id="Bwrite" value="30MBps" />
833 <model_prop id="Bread" value="100MBps" />
834 <model_prop id="Bconnection" value="150MBps" />
835 <prop id="Brand" value="Western Digital" />
839 \anchor pf_storage_entity_storage
840 #### <storage> ####
842 ``storage`` attributes:
844 Attribute name | Mandatory | Values | Description
845 -------------- | --------- | ------ | -----------
846 id | yes | string | Identifier of this ``storage``; used when referring to it
847 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.
848 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)
849 content | no | string | When specified, overwrites the content attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
850 content_type | no | string | When specified, overwrites the content_type attribute of \ref pf_storage_entity_storage_type "\<storage_type\>"
852 Here are two examples:
855 <storage id="Disk1" typeId="single_HDD" attach="bob" />
857 <storage id="Disk2" typeId="single_SSD"
858 content="content/win_storage_content.txt"
859 content_type="txt_windows" attach="alice" />
862 The first example is straightforward: A disk is defined and called "Disk1"; it is
863 of type "single_HDD" (shown as an example of \ref pf_storage_entity_storage_type "\<storage_type\>" above) and attached
864 to a host called "bob" (the definition of this host is omitted here).
866 The second storage is called "Disk2", is still of the same type as Disk1 but
867 now specifies a new content file (so the contents will be different from Disk1)
868 and the filesystem uses the windows style; finally, it is attached to a second host,
869 called alice (which is again not defined here).
871 \anchor pf_storage_entity_mount
872 #### <mount> ####
875 | Attribute name | Mandatory | Values | Description |
876 | ---------------- | ----------- | -------- | ------------- |
877 | id | yes | string | Refers to a \ref pf_storage_entity_storage "<storage>" entity that will be mounted on that computer |
878 | name | yes | string | Path/location to/of the logical reference (mount point) of this disk
880 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)
881 is; this location is specified by the ``name`` attribute.
883 Here is a simple example, taken from the file ``examples/platform/storage.xml``:
886 <storage_type id="single_SSD" model="linear_no_lat" size="500GiB">
887 <model_prop id="Bwrite" value="60MBps" />
888 <model_prop id="Bread" value="200MBps" />
889 <model_prop id="Bconnection" value="220MBps" />
892 <storage id="Disk2" typeId="single_SSD"
893 content="content/win_storage_content.txt"
894 content_type="txt_windows" attach="alice" />
895 <storage id="Disk4" typeId="single_SSD"
896 content="content/small_content.txt"
897 content_type="txt_unix" attach="denise"/>
899 <host id="alice" power="1Gf">
900 <mount storageId="Disk2" name="c:"/>
903 <host id="denise" power="1Gf">
904 <mount storageId="Disk2" name="c:"/>
905 <mount storageId="Disk4" name="/home"/>
909 This example is quite interesting, as the same device, called "Disk2", is mounted by
910 two hosts at the same time! Note, however, that the host called ``alice`` is actually
911 attached to this storage, as can be seen in the \ref pf_storage_entity_storage "<storage>"
912 tag. This means that ``denise`` must access this storage through the network, but SimGrid automatically takes
913 care of that for you.
915 Furthermore, this example shows that ``denise`` has mounted two storages with different
916 filesystem types (unix and windows). In general, a host can mount as many storage devices as
920 Again, the difference between ``attach`` and ``mount`` is simply that
921 an attached storage is always physically inside (or connected to) that machine;
922 for instance, a USB stick is attached to one and only one machine (where it's plugged-in)
923 but it can only be mounted on others, as mounted storage can also be a remote location.
925 ###### Example files #####
927 \verbinclude example_filelist_xmltag_mount
929 \subsubsection pf_storage_example_files Example files
931 Several examples were already discussed above; if you're interested in full examples,
932 check the the following platforms:
934 1. ``examples/platforms/storage.xml``
935 2. ``examples/platforms/remote_io.xml``
937 If you're looking for some examplary C code, you may find the source code
938 available in the directory ``examples/msg/io/`` useful.
940 \subsubsection pf_storage_examples_modelling Modelling different situations
942 The storage functionality of SimGrid is type-agnostic, that is, the implementation
943 does not presume any type of storage, such as HDDs/SSDs, RAM,
944 CD/DVD devices, USB sticks etc.
946 This allows the user to apply the simulator for a wide variety of scenarios; one
947 common scenario would be the access of remote RAM.
949 #### Modelling the access of remote RAM ####
951 How can this be achieved in SimGrid? Let's assume we have a setup where three hosts
952 (HostA, HostB, HostC) need to access remote RAM:
962 An easy way to model this scenario is to setup and define the RAM via the
963 \ref pf_storage_entity_storage "storage" and \ref pf_storage_entity_storage_type "storage type"
964 entities and attach it to a remote dummy host; then, every host can have their own links
965 to this host (modelling for instance certain scenarios, such as PCIe ...)
970 RAM - Dummy -- Host B
975 Now, if read from this storage, the host that mounts this storage
976 communicates to the dummy host which reads from RAM and
977 sends the information back.
980 \section pf_routing Routing
982 To achieve high performance, the routing tables used within SimGrid are
983 static. This means that routing between two nodes is calculated once
984 and will not change during execution. The SimGrid team chose to use this
985 approach as it is rare to have a real deficiency of a resource;
986 most of the time, a communication fails because the links experience too much
987 congestion and hence, your connection stops before the timeout or
988 because the computer designated to be the destination of that message
991 We also chose to use shortest paths algorithms in order to emulate
992 routing. Doing so is consistent with the reality: [RIP](https://en.wikipedia.org/wiki/Routing_Information_Protocol),
993 [OSPF](https://en.wikipedia.org/wiki/Open_Shortest_Path_First), [BGP](https://en.wikipedia.org/wiki/Border_Gateway_Protocol)
994 are all calculating shortest paths. They do require some time to converge, but
995 eventually, when the routing tables have stabilized, your packets will follow
998 \subsection pf_rm Routing models
1000 For each AS, you must define explicitly which routing model will
1001 be used. There are 3 different categories for routing models:
1003 1. \ref pf_routing_model_shortest_path "Shortest-path" based models: SimGrid calculates shortest
1004 paths and manages them. Behaves more or less like most real life
1006 2. \ref pf_routing_model_manual "Manually-entered" route models: you have to define all routes
1007 manually in the platform description file; this can become
1008 tedious very quickly, as it is very verbose.
1009 Consistent with some manually managed real life routing.
1010 3. \ref pf_routing_model_simple "Simple/fast models": those models offer fast, low memory routing
1011 algorithms. You should consider to use this type of model if
1012 you can make some assumptions about your AS.
1013 Routing in this case is more or less ignored.
1015 \subsubsection pf_raf The router affair
1017 Using routers becomes mandatory when using shortest-path based
1018 models or when using the bindings to the ns-3 packet-level
1019 simulator instead of the native analytical network model implemented
1022 For graph-based shortest path algorithms, routers are mandatory, because these
1023 algorithms require a graph as input and so we need to have source and
1024 destination for each edge.
1026 Routers are naturally an important concept ns-3 since the
1027 way routers run the packet routing algorithms is actually simulated.
1028 SimGrid's analytical models however simply aggregate the routing time
1029 with the transfer time.
1031 So why did we incorporate routers in SimGrid? Rebuilding a graph representation
1032 only from the route information turns out to be a very difficult task, because
1033 of the missing information about how routes intersect. That is why we
1034 introduced routers, which are simply used to express these intersection points.
1035 It is important to understand that routers are only used to provide topological
1038 To express this topological information, a <b>route</b> has to be
1039 defined in order to declare which link is connected to a router.
1042 \subsubsection pf_routing_model_shortest_path Shortest-path based models
1044 The following table shows all the models that compute routes using
1045 shortest-paths algorithms are currently available in SimGrid. More detail on how
1046 to choose the best routing model is given in the Section called \"\ref pf_routing_howto_choose_wisely\".
1048 | Name | Description |
1049 | --------------------------------------------------- | -------------------------------------------------------------------------- |
1050 | \ref pf_routing_model_floyd "Floyd" | Floyd routing data. Pre-calculates all routes once |
1051 | \ref pf_routing_model_dijkstra "Dijkstra" | Dijkstra routing data. Calculates routes only when needed |
1052 | \ref pf_routing_model_dijkstracache "DijkstraCache" | Dijkstra routing data. Handles some cache for already calculated routes. |
1054 All those shortest-path models are instanciated in the same way and are
1055 completely interchangeable. Here are some examples:
1057 \anchor pf_routing_model_floyd
1062 <AS id="AS0" routing="Floyd">
1064 <cluster id="my_cluster_1" prefix="c-" suffix=""
1065 radical="0-1" power="1000000000" bw="125000000" lat="5E-5"
1066 router_id="router1"/>
1068 <AS id="AS1" routing="None">
1069 <host id="host1" power="1000000000"/>
1072 <link id="link1" bandwidth="100000" latency="0.01"/>
1074 <ASroute src="my_cluster_1" dst="AS1"
1077 <link_ctn id="link1"/>
1083 ASroute given at the end gives a topological information: link1 is
1084 between router1 and host1.
1086 #### Example platform files ####
1088 This is an automatically generated list of example files that use the Floyd
1089 routing model (the path is given relative to SimGrid's source directory)
1091 \verbinclude example_filelist_routing_floyd
1093 \anchor pf_routing_model_dijkstra
1096 #### Example platform files ####
1098 This is an automatically generated list of example files that use the Dijkstra
1099 routing model (the path is given relative to SimGrid's source directory)
1101 \verbinclude example_filelist_routing_dijkstra
1105 <AS id="AS_2" routing="Dijkstra">
1106 <host id="AS_2_host1" power="1000000000"/>
1107 <host id="AS_2_host2" power="1000000000"/>
1108 <host id="AS_2_host3" power="1000000000"/>
1109 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1110 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1111 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1112 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1113 <router id="central_router"/>
1114 <router id="AS_2_gateway"/>
1115 <!-- routes providing topological information -->
1116 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1117 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1118 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1119 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1123 \anchor pf_routing_model_dijkstracache
1124 ### DijkstraCache ###
1126 DijkstraCache example:
1128 <AS id="AS_2" routing="DijkstraCache">
1129 <host id="AS_2_host1" power="1000000000"/>
1131 (platform unchanged compared to upper example)
1134 #### Example platform files ####
1136 This is an automatically generated list of example files that use the DijkstraCache
1137 routing model (the path is given relative to SimGrid's source directory):
1139 Editor's note: At the time of writing, no platform file used this routing model - so
1140 if there are no example files listed here, this is likely to be correct.
1142 \verbinclude example_filelist_routing_dijkstra_cache
1144 \subsubsection pf_routing_model_manual Manually-entered route models
1146 | Name | Description |
1147 | ---------------------------------- | ------------------------------------------------------------------------------ |
1148 | \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. |
1150 \anchor pf_routing_model_full
1155 <AS id="AS0" routing="Full">
1156 <host id="host1" power="1000000000"/>
1157 <host id="host2" power="1000000000"/>
1158 <link id="link1" bandwidth="125000000" latency="0.000100"/>
1159 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
1163 #### Example platform files ####
1165 This is an automatically generated list of example files that use the Full
1166 routing model (the path is given relative to SimGrid's source directory):
1168 \verbinclude example_filelist_routing_full
1170 \subsubsection pf_routing_model_simple Simple/fast models
1172 | Name | Description |
1173 | ---------------------------------------- | ------------------------------------------------------------------------------ |
1174 | \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. |
1175 | \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. |
1176 | \ref pf_routing_model_vivaldi "Vivaldi" | Perfect when you want to use coordinates. Also see the corresponding \ref pf_P2P_tags "P2P section" below. |
1178 \anchor pf_routing_model_cluster
1182 In this mode, the \ref pf_cabinet "<cabinet/>" tag is available.
1184 #### Example platform files ####
1186 This is an automatically generated list of example files that use the Cluster
1187 routing model (the path is given relative to SimGrid's source directory):
1189 \verbinclude example_filelist_routing_cluster
1191 \anchor pf_routing_model_none
1194 This model does exactly what it's name advertises: Nothing. There is no routing
1195 available within this model and if you try to communicate within the AS that
1196 uses this model, SimGrid will fail unless you have explicitly activated the
1197 \ref options_model_select_network_constant "Constant Network Model" (this model charges
1198 the same for every single communication). It should
1199 be noted, however, that you can still attach an \ref pf_routing_tag_asroute "ASroute",
1200 as is demonstrated in the example below:
1202 \verbinclude platforms/cluster_and_one_host.xml
1204 #### Example platform files ####
1206 This is an automatically generated list of example files that use the None
1207 routing model (the path is given relative to SimGrid's source directory):
1209 \verbinclude example_filelist_routing_none
1212 \anchor pf_routing_model_vivaldi
1215 For more information on how to use the [Vivaldi Coordinates](https://en.wikipedia.org/wiki/Vivaldi_coordinates),
1216 see also Section \ref pf_P2P_tags "P2P tags".
1218 For documentation on how to activate this model (as some initialization must be done
1219 in the simulator), see Section \ref options_model_network_coord "Activating Coordinate Based Routing".
1221 Note that it is possible to combine the Vivaldi routing model with other routing models;
1222 an example can be found in the file \c examples/platforms/cloud.xml. This
1223 examples models an AS using Vivaldi that contains other ASes that use different
1226 #### Example platform files ####
1228 This is an automatically generated list of example files that use the None
1229 routing model (the path is given relative to SimGrid's source directory):
1231 \verbinclude example_filelist_routing_vivaldi
1234 \subsection ps_dec Defining routes
1236 There are currently four different ways to define routes:
1238 | Name | Description |
1239 | ------------------------------------------------- | ----------------------------------------------------------------------------------- |
1240 | \ref pf_routing_tag_route "route" | Used to define route between host/router |
1241 | \ref pf_routing_tag_asroute "ASroute" | Used to define route between different AS |
1242 | \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. |
1243 | \ref pf_routing_tag_bypassasroute "bypassASroute" | Used in the same way as bypassRoute, but for AS |
1245 Basically all those tags will contain an (ordered) list of references
1246 to link that compose the route you want to define.
1248 Consider the example below:
1251 <route src="Alice" dst="Bob">
1252 <link_ctn id="link1"/>
1253 <link_ctn id="link2"/>
1254 <link_ctn id="link3"/>
1258 The route here from host Alice to Bob will be first link1, then link2,
1259 and finally link3. What about the reverse route? \ref pf_routing_tag_route "Route" and
1260 \ref pf_routing_tag_asroute "ASroute" have an optional attribute \c symmetrical, that can
1261 be either \c YES or \c NO. \c YES means that the reverse route is the same
1262 route in the inverse order, and is set to \c YES by default. Note that
1263 this is not the case for bypass*Route, as it is more probable that you
1264 want to bypass only one default route.
1266 For an \ref pf_routing_tag_asroute "ASroute", things are just slightly more complicated, as you have
1267 to give the id of the gateway which is inside the AS you want to access ...
1268 So it looks like this:
1271 <ASroute src="AS1" dst="AS2"
1272 gw_src="router1" gw_dst="router2">
1273 <link_ctn id="link1"/>
1277 gw == gateway, so when any message are trying to go from AS1 to AS2,
1278 it means that it must pass through router1 to get out of the AS, then
1279 pass through link1, and get into AS2 by being received by router2.
1280 router1 must belong to AS1 and router2 must belong to AS2.
1282 \subsubsection pf_linkctn <link_ctn/>
1284 This entity has only one purpose: Refer to an already existing
1285 \ref pf_link "<link/>" when defining a route, i.e., it
1286 can only occur as a child of \ref pf_routing_tag_route "<route/>"
1288 | Attribute name | Mandatory | Values | Description |
1289 | --------------- | --------- | ------ | ----------- |
1290 | id | yes | String | The identifier of the link that should be added to the route. |
1291 | 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.
1293 #### Example Files ####
1295 This is an automatically generated list of example files that use the \c <link_ctn/>
1296 entity (the path is given relative to SimGrid's source directory):
1298 \verbinclude example_filelist_xmltag_linkctn
1300 \subsubsection pf_routing_tag_asroute ASroute
1302 The purpose of this entity is to define a route between two ASes.
1303 This is mainly useful when you're in the \ref pf_routing_model_full "Full routing model".
1305 #### Attributes ####
1307 | Attribute name | Mandatory | Values | Description |
1308 | --------------- | --------- | ------ | ----------- |
1309 | src | yes | String | The identifier of the source AS |
1310 | dst | yes | String | See the \c src attribute |
1311 | 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. |
1312 | gw_dst | yes | String | Same as \c gw_src, but with the dst AS instead. |
1313 | symmetrical | no | YES\|NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1318 <AS id="AS0" routing="Full">
1319 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
1320 radical="0-149" power="1000000000" bw="125000000" lat="5E-5"
1321 bb_bw="2250000000" bb_lat="5E-4"/>
1323 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
1324 radical="150-299" power="1000000000" bw="125000000" lat="5E-5"
1325 bb_bw="2250000000" bb_lat="5E-4"/>
1327 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1329 <ASroute src="my_cluster_1" dst="my_cluster_2"
1330 gw_src="c-my_cluster_1_router.me"
1331 gw_dst="c-my_cluster_2_router.me">
1332 <link_ctn id="backbone"/>
1334 <ASroute src="my_cluster_2" dst="my_cluster_1"
1335 gw_src="c-my_cluster_2_router.me"
1336 gw_dst="c-my_cluster_1_router.me">
1337 <link_ctn id="backbone"/>
1342 \subsubsection pf_routing_tag_route route
1344 The principle is the same as for
1345 \ref pf_routing_tag_asroute "ASroute": The route contains a list of links that
1346 provide a path from \c src to \c dst. Here, \c src and \c dst can both be either a
1347 \ref pf_host "host" or \ref pf_router "router". This is mostly useful for the
1348 \ref pf_routing_model_full "Full routing model" as well as for the
1349 \ref pf_routing_model_shortest_path "shortest-paths" based models (as they require
1350 topological information).
1353 | Attribute name | Mandatory | Values | Description |
1354 | --------------- | --------- | ---------------------- | ----------- |
1355 | src | yes | String | The value given to the source's "id" attribute |
1356 | dst | yes | String | The value given to the destination's "id" attribute. |
1357 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1362 A route in the \ref pf_routing_model_full "Full routing model" could look like this:
1364 <route src="Tremblay" dst="Bourassa">
1365 <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"/>
1369 A route in the \ref pf_routing_model_shortest_path "Shortest-Path routing model" could look like this:
1371 <route src="Tremblay" dst="Bourassa">
1376 You must only have one link in your routes when you're using them to provide
1377 topological information, as the routes here are simply the edges of the
1378 (network-)graph and the employed algorithms need to know which edge connects
1379 which pair of entities.
1381 \subsubsection pf_routing_tag_bypassasroute bypassASroute
1383 %As said before, once you choose
1384 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1385 define some of your routes, which will be specific. You may also want
1386 to bypass some routes defined in lower level AS at an upper stage:
1387 <b>bypassASroute</b> is the tag you're looking for. It allows to
1388 bypass routes defined between already defined between AS (if you want
1389 to bypass route for a specific host, you should just use byPassRoute).
1390 The principle is the same as ASroute : <b>bypassASroute</b> contains
1391 list of links that are in the path between src and dst.
1393 #### Attributes ####
1395 | Attribute name | Mandatory | Values | Description |
1396 | --------------- | --------- | ---------------------- | ----------- |
1397 | src | yes | String | The value given to the source AS's "id" attribute |
1398 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1399 | 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 |
1400 | 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|
1401 | symmetrical | no | YES\| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1406 <bypassASRoute src="my_cluster_1" dst="my_cluster_2"
1407 gw_src="my_cluster_1_router"
1408 gw_dst="my_cluster_2_router">
1409 <link_ctn id="link_tmp"/>
1413 This example shows that link \c link_tmp (definition not displayed here) directly
1414 connects the router \c my_cluster_1_router in the source cluster to the router
1415 \c my_cluster_2_router in the destination router. Additionally, as the \c symmetrical
1416 attribute was not given, this route is presumed to be symmetrical.
1418 \subsubsection pf_routing_tag_bypassroute bypassRoute
1420 %As said before, once you choose
1421 a model, it (most likely; the constant network model, for example, doesn't) calculates routes for you. But maybe you want to
1422 define some of your routes, which will be specific. You may also want
1423 to bypass some routes defined in lower level AS at an upper stage :
1424 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
1425 routes defined between <b>host/router</b>. The principle is the same
1426 as route : <b>bypassRoute</b> contains list of links references of
1427 links that are in the path between src and dst.
1429 #### Attributes ####
1431 | Attribute name | Mandatory | Values | Description |
1432 | --------------- | --------- | ---------------------- | ----------- |
1433 | src | yes | String | The value given to the source AS's "id" attribute |
1434 | dst | yes | String | The value given to the destination AS's "id" attribute. |
1435 | symmetrical | no | YES \| NO (Default: YES) | If this route is symmetric, the opposite route (from dst to src) will also be declared implicitly. |
1440 <bypassRoute src="host_1" dst="host_2">
1441 <link_ctn id="link_tmp"/>
1445 This example shows that link \c link_tmp (definition not displayed here) directly
1446 connects host \c host_1 to host \c host_2. Additionally, as the \c symmetrical
1447 attribute was not given, this route is presumed to be symmetrical.
1449 \subsection pb_baroex Basic Routing Example
1451 Let's say you have an AS named AS_Big that contains two other AS, AS_1
1452 and AS_2. If you want to make a host (h1) from AS_1 with another one
1453 (h2) from AS_2 then you'll have to proceed as follows:
1454 \li First, you have to ensure that a route is defined from h1 to the
1455 AS_1's exit gateway and from h2 to AS_2's exit gateway.
1456 \li Then, you'll have to define a route between AS_1 to AS_2. %As those
1457 AS are both resources belonging to AS_Big, then it has to be done
1458 at AS_big level. To define such a route, you have to give the
1459 source AS (AS_1), the destination AS (AS_2), and their respective
1460 gateway (as the route is effectively defined between those two
1461 entry/exit points). Elements of this route can only be elements
1462 belonging to AS_Big, so links and routers in this route should be
1463 defined inside AS_Big. If you choose some shortest-path model,
1464 this route will be computed automatically.
1466 %As said before, there are mainly 2 tags for routing :
1467 \li <b>ASroute</b>: to define routes between two <b>AS</b>
1468 \li <b>route</b>: to define routes between two <b>host/router</b>
1470 %As we are dealing with routes between AS, it means that those we'll
1471 have some definition at AS_Big level. Let consider AS_1 contains 1
1472 host, 1 link and one router and AS_2 3 hosts, 4 links and one router.
1473 There will be a central router, and a cross-like topology. At the end
1474 of the crosses arms, you'll find the 3 hosts and the router that will
1475 act as a gateway. We have to define routes inside those two AS. Let
1476 say that AS_1 contains full routes, and AS_2 contains some Floyd
1477 routing (as we don't want to bother with defining all routes). %As
1478 we're using some shortest path algorithms to route into AS_2, we'll
1479 then have to define some <b>route</b> to gives some topological
1480 information to SimGrid. Here is a file doing it all :
1483 <AS id="AS_Big" routing="Dijkstra">
1484 <AS id="AS_1" routing="Full">
1485 <host id="AS_1_host1" power="1000000000"/>
1486 <link id="AS_1_link" bandwidth="1250000000" latency="5E-4"/>
1487 <router id="AS_1_gateway"/>
1488 <route src="AS_1_host1" dst="AS_1_gateway">
1489 <link_ctn id="AS_1_link"/>
1492 <AS id="AS_2" routing="Floyd">
1493 <host id="AS_2_host1" power="1000000000"/>
1494 <host id="AS_2_host2" power="1000000000"/>
1495 <host id="AS_2_host3" power="1000000000"/>
1496 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1497 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1498 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1499 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1500 <router id="central_router"/>
1501 <router id="AS_2_gateway"/>
1502 <!-- routes providing topological information -->
1503 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1504 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1505 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1506 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1508 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1510 <ASroute src="AS_1" dst="AS_2"
1511 gw_src="AS_1_gateway"
1512 gw_dst="AS_2_gateway">
1513 <link_ctn id="backbone"/>
1518 \section pf_other_tags Tags not (directly) describing the platform
1520 The following tags can be used inside a \<platform\> tag even if they are not
1521 directly describing the platform:
1522 \li \ref pf_config "config": it allows you to pass some configuration stuff like, for
1523 example, the network model and so on. It follows the
1524 \li \ref pf_include "include": allows you to include another file into the current one.
1526 \subsection pf_config config
1528 The only purpose of this tag is to contain the \c prop tags, as described below.
1529 These tags will then configure the options as described by Section \ref options.
1532 #### Attributes ####
1534 | Attribute name | Mandatory | Values | Description |
1535 | --------------- | --------- | ---------------------- | ----------- |
1536 | id | yes | String | The identifier of the config tag when referring to id; this is basically useless, though. |
1538 #### Possible children ####
1540 Tag name | Description | Documentation
1541 ------------ | ----------- | -------------
1542 \<prop/\> | The prop tag allows you to define different configuration options following the attribute/value schema. See the \ref options page. | N/A
1547 <?xml version='1.0'?>
1548 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1549 <platform version="4">
1550 <config id="General">
1551 <prop id="maxmin/precision" value="0.000010"></prop>
1552 <prop id="cpu/optim" value="TI"></prop>
1553 <prop id="host/model" value="compound"></prop>
1554 <prop id="network/model" value="SMPI"></prop>
1555 <prop id="path" value="~/"></prop>
1556 <prop id="smpi/bw_factor" value="65472:0.940694;15424:0.697866;9376:0.58729"></prop>
1559 <AS id="AS0" routing="Full">
1563 \subsection pf_include include
1565 Even if it can be used in other contexts, this tag was originally created
1566 to be used with \ref pf_trace. The idea was to have a file describing the
1567 platform, and another file attaching traces of a given period to the platform.
1569 The drawback is that the file chuncks that will be included do not
1570 constitute valid XML files. This may explain why this feature was never really
1571 used in practice (as far as we know). Other mechanisms, such as the ability to load
1572 several platform files one after the other, could be considered in the future.
1574 In the meanwhile, the \c include tag allows you to import other platforms into your
1575 local file. This is done with the intention to help people
1576 combine their different AS and provide new platforms. Those files
1577 should contain XML that consists of
1578 \ref pf_include "include", \ref pf_cluster "cluster", \ref pf_peer "peer", \ref pf_As "AS", \ref pf_trace "trace", \ref pf_trace "tags".
1580 Do not forget to close the tag to make it work, or you will end up with an invalid XML file.
1582 #### Attributes ####
1584 | Attribute name | Mandatory | Values | Description |
1585 | --------------- | --------- | ---------------------- | ----------- |
1586 | file | yes | String | Filename of the path you want to include with either relative or absolute path. |
1591 The following example includes two files, clusterA.xml and clusterB.xml and
1592 combines them two one platform file; all hosts, routers etc. defined in
1593 each of them will then be usable.
1596 <?xml version='1.0'?>
1597 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid/simgrid.dtd">
1598 <platform version="4">
1599 <AS id="main" routing="Full">
1600 <include file="clusterA.xml"></include>
1601 <include file="clusterB.xml"></include>
1606 \subsection pf_trace trace and trace_connect
1608 Both tags are an alternate way to pass files containing information on
1609 availability, state etc. to an entity. (See also, for instance, Section \ref
1610 pf_host_churn "Churn", as described for the host entity.) Instead of referring
1611 to the file directly in the host, link, or cluster tag, you proceed by defining
1612 a trace with an id corresponding to a file, later a host/link/cluster, and
1613 finally using trace_connect you say that the file trace must be used by the
1620 <AS id="AS0" routing="Full">
1621 <host id="bob" power="1000000000"/>
1623 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1624 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1628 The order here is important. \c trace_connect must come
1629 after the elements \c trace and \c host, as both the host
1630 and the trace definition must be known when \c trace_connect
1631 is parsed; the order of \c trace and \c host is arbitrary.
1634 #### \c trace attributes ####
1637 | Attribute name | Mandatory | Values | Description |
1638 | --------------- | --------- | ---------------------- | ----------- |
1639 | id | yes | String | Identifier of this trace; this is the name you pass on to \c trace_connect. |
1640 | 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. |
1641 | trace_periodicity | yes | String | This is the same as for \ref pf_host "hosts" (see there for details) |
1643 Here is an example of trace when no file name is provided:
1646 <trace id="myTrace" periodicity="1.0">
1653 #### \c trace_connect attributes ####
1655 | Attribute name | Mandatory | Values | Description |
1656 | --------------- | --------- | ---------------------- | ----------- |
1657 | kind | no | HOST_AVAIL\|POWER\|<br/>LINK_AVAIL\|BANDWIDTH\|LATENCY (Default: HOST_AVAIL) | Describes the kind of trace. |
1658 | trace | yes | String | Identifier of the referenced trace (specified of the trace's \c id attribute) |
1659 | element | yes | String | The identifier of the referenced entity as given by its \c id attribute |
1661 \section pf_hints Hints and tips, or how to write a platform efficiently
1663 Now you should know at least the syntax and be able to create a
1664 platform by your own. However, after having ourselves wrote some platforms, there
1665 are some best practices you should pay attention to in order to
1666 produce good platform and some choices you can make in order to have
1667 faster simulations. Here's some hints and tips, then.
1669 \subsection pf_as_h AS Hierarchy
1670 The AS design allows SimGrid to go fast, because computing route is
1671 done only for the set of resources defined in this AS. If you're using
1672 only a big AS containing all resource with no AS into it and you're
1673 using Full model, then ... you'll loose all interest into it. On the
1674 other hand, designing a binary tree of AS with, at the lower level,
1675 only one host, then you'll also loose all the good AS hierarchy can
1676 give you. Remind you should always be "reasonable" in your platform
1677 definition when choosing the hierarchy. A good choice if you try to
1678 describe a real life platform is to follow the AS described in
1679 reality, since this kind of trade-off works well for real life
1682 \subsection pf_exit_as Exit AS: why and how
1683 Users that have looked at some of our platforms may have notice a
1684 non-intuitive schema ... Something like that :
1688 <AS id="AS_4" routing="Full">
1689 <AS id="exitAS_4" routing="Full">
1690 <router id="router_4"/>
1692 <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"/>
1693 <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"/>
1694 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1695 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1696 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1697 <ASroute src="cl_4_1"
1699 gw_src="c_4_1-cl_4_1_router"
1700 gw_dst="c_4_2-cl_4_2_router">
1701 <link_ctn id="4_1"/>
1702 <link_ctn id="bb_4"/>
1703 <link_ctn id="4_2"/>
1705 <ASroute src="cl_4_1"
1707 gw_src="c_4_1-cl_4_1_router"
1709 <link_ctn id="4_1"/>
1710 <link_ctn id="bb_4"/>
1712 <ASroute src="cl_4_2"
1714 gw_src="c_4_2-cl_4_2_router"
1716 <link_ctn id="4_2"/>
1717 <link_ctn id="bb_4"/>
1722 In the AS_4, you have an exitAS_4 defined, containing only one router,
1723 and routes defined to that AS from all other AS (as cluster is only a
1724 shortcut for an AS, see cluster description for details). If there was
1725 an upper AS, it would define routes to and from AS_4 with the gateway
1726 router_4. It's just because, as we did not allowed (for performances
1727 issues) to have routes from an AS to a single host/router, you have to
1728 enclose your gateway, when you have AS included in your AS, within an
1729 AS to define routes to it.
1731 \subsection pf_P2P_tags P2P or how to use coordinates
1732 SimGrid allows you to use some coordinated-based system, like vivaldi,
1733 to describe a platform. The main concept is that you have some peers
1734 that are located somewhere: this is the function of the
1735 <b>coordinates</b> of the \<peer\> or \<host\> tag. There's nothing
1736 complicated in using it, here is an example of it:
1739 <?xml version='1.0'?>
1740 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1741 <platform version="4">
1743 <config id="General">
1744 <prop id="network/coordinates" value="yes"></prop>
1746 <AS id="AS0" routing="Vivaldi">
1747 <host id="100030591" coordinates="25.5 9.4 1.4" power="1.5Gf" />
1748 <host id="100036570" coordinates="-12.7 -9.9 2.1" power="7.3Gf" />
1750 <host id="100429957" coordinates="17.5 6.7 18.8" power="8.3Gf" />
1755 Coordinates are then used to calculate latency (in microseconds)
1756 between two hosts by calculating the distance between the two hosts
1757 coordinates with the following formula: distance( (x1, y1, z1), (x2,
1758 y2, z2) ) = euclidian( (x1,y1), (x2,y2) ) + abs(z1) + abs(z2)
1760 In other words, we take the euclidian distance on the two first
1761 dimensions, and then add the absolute values found on the third
1762 dimension. This may seem strange, but it was found to allow better
1763 approximations of the latency matrices (see the paper describing
1766 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between AS.
1767 That is for example what is commonly done when using peers (see Section \ref pf_peer).
1769 <?xml version='1.0'?>
1770 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1771 <platform version="4">
1773 <config id="General">
1774 <prop id="network/coordinates" value="yes"></prop>
1776 <AS id="AS0" routing="Vivaldi">
1777 <peer id="peer-0" coordinates="173.0 96.8 0.1" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1778 <peer id="peer-1" coordinates="247.0 57.3 0.6" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1779 <peer id="peer-2" coordinates="243.4 58.8 1.4" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1783 In such a case though, we connect the AS created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1784 This means that to route between AS1 and AS2, it will use the coordinates of router_AS1 and router_AS2.
1785 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1786 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.
1789 \subsection pf_routing_howto_choose_wisely Choosing wisely the routing model to use
1792 Choosing wisely the routing model to use can significantly fasten your
1793 simulation/save your time when writing the platform/save tremendous
1794 disk space. Here is the list of available model and their
1795 characteristics (lookup : time to resolve a route):
1797 \li <b>Full</b>: Full routing data (fast, large memory requirements,
1799 \li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1800 lookup, lesser memory requirements, shortest path routing only).
1801 Calculates all routes at once at the beginning.
1802 \li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1803 lookup, small memory requirements, shortest path routing only).
1804 Calculates a route when necessary.
1805 \li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1806 fast lookup, small memory requirements, shortest path routing
1807 only). Same as Dijkstra, except it handles a cache for latest used
1809 \li <b>None</b>: No routing (usable with Constant network only).
1810 Defines that there is no routes, so if you try to determine a
1811 route without constant network within this AS, SimGrid will raise
1813 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1814 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1817 \subsection pf_switch How to describe a switch given that is no switch tag?
1819 Actually we did not include switch tag, ok. But when you're trying to
1820 simulate a switch, the only major impact it has when you're using
1821 fluid model (and SimGrid uses fluid model unless you activate
1822 ns-3 or constant network mode) is the impact of the upper limit of
1823 the switch motherboard speed that will eventually be reached if you're
1824 using intensively your switch. So, the switch impact is similar to a
1825 link one. That's why we are used to describe a switch using a link tag
1826 (as a link is not an edge by a hyperedge, you can connect more than 2
1829 \subsection pf_multicabinets How to model multi-cabinets clusters?
1831 You have several possibilities, as usual when modeling things. If your
1832 cabinets are homogeneous and the intercabinet network negligible for
1833 your study, you should just create a larger cluster with all hosts at
1836 In the rare case where your hosts are not homogeneous between the
1837 cabinets, you can create your cluster completely manually. For that,
1838 create an As using the Cluster routing, and then use one
1839 <cabinet> for each cabinet. This cabinet tag can only be used an
1840 As using the Cluster routing schema, and creating
1842 Be warned that creating a cluster manually from the XML with
1843 <cabinet>, <backbone> and friends is rather tedious. The
1844 easiest way to retrieve some control of your model without diving into
1845 the <cluster> internals is certainly to create one separate
1846 <cluster> per cabinet and interconnect them together. This is
1847 what we did in the G5K example platform for the Graphen cluster.
1849 \subsection pf_platform_multipath How to express multipath routing in platform files?
1851 It is unfortunately impossible to express the fact that there is more
1852 than one routing path between two given hosts. Let's consider the
1853 following platform file:
1856 <route src="A" dst="B">
1859 <route src="B" dst="C">
1862 <route src="A" dst="C">
1867 Although it is perfectly valid, it does not mean that data traveling
1868 from A to C can either go directly (using link 3) or through B (using
1869 links 1 and 2). It simply means that the routing on the graph is not
1870 trivial, and that data do not following the shortest path in number of
1871 hops on this graph. Another way to say it is that there is no implicit
1872 in these routing descriptions. The system will only use the routes you
1873 declare (such as <route src="A" dst="C"><link_ctn
1874 id="3"/></route>), without trying to build new routes by aggregating
1877 You are also free to declare platform where the routing is not
1878 symmetrical. For example, add the following to the previous file:
1881 <route src="C" dst="A">
1887 This makes sure that data from C to A go through B where data from A
1888 to C go directly. Don't worry about realism of such settings since
1889 we've seen ways more weird situation in real settings (in fact, that's
1890 the realism of very regular platforms which is questionable, but
1891 that's another story).
