1 /*! \page platform Platform Description
5 In order to run any simulation, SimGrid needs 3 things: something to run
6 (so, your code), a description of the platform on which you want to run your
7 application, and finally it needs something to know where to deploy what.
9 For the latest 2 entries, you have basically 2 ways to give it as an input :
10 \li You can program it, either using the Lua console (\ref
11 MSG_Lua_funct) or if you're using MSG some of its platform and
12 deployments functions(\ref msg_simulation). If you want to use it,
13 please refer to its doc. (you can also check the section \ref
14 pf_flexml_bypassing but this is strongly deprecated, as there is a
15 new way to do it properly, but not yet documented).
16 \li You can use two XML files: a platform description file and a
17 deployment description one.
19 For the deployment stuff, please take a look at \ref deployment
21 The platform description may be complicated. This documentation is all
22 about how to write this file: what are the basic concept it relies on,
23 what possibilities are offered, and some hints and tips on how to
24 write a good platform description.
26 \section pf_overview Some words about XML and DTD
28 We choose to use XML because of some of its possibilities: if you're
29 using an accurate XML editor, or simply using any XML plug-in for
30 eclipse, it will allow you to have cool stuff like auto-completion,
31 validation and checking, so all syntax errors may be avoided this
34 the XML checking is done based on the dtd which is nowadays online at
35 <a href="http://simgrid.gforge.inria.fr/simgrid.dtd">http://simgrid.gforge.inria.fr/simgrid.dtd</a>
36 while you might be tempted to read it, it will not help you that much.
38 If you read it, you should notice two or three important things :
39 \li The platform tags contains a version attributes. At the time of
40 writing this doc the current version is 3.
41 \li The DTD contains definitions for the 2 files used by SimGrid (platform
42 description and deployment).
43 \li There is a bunch of possibilities ! Let's see what's in it
46 \section pf_basics Basic concepts
48 Nowadays, the Internet is composed of a bunch of independently managed
49 networks. Within each of those networks, there are entry and exit
50 points (most of the time, you can both enter and exit through the same
51 point) that allows to go out of the current network and reach other
52 networks. At the upper level, these networks are known as
53 <b>Autonomous System (AS)</b>, while at the lower level they are named
54 sub-networks, or LAN. Indeed they are autonomous: routing is defined
55 within the limits of his network by the administrator, and so, those
56 networks can continue to operate without the existence of other
57 networks. There are some rules to get out of networks by the entry
58 points (or gateways). Those gateways allow you to go from a network to
59 another one. Inside of each autonomous system, there is a bunch of
60 equipments (cables, routers, switches, computers) that belong to the
61 autonomous system owner.
63 SimGrid platform description file relies exactly on the same concepts
64 as real life platform. Every resource (computers, network equipments,
65 and so on) belongs to an AS. Within this AS, you can define the
66 routing you want between its elements (that's done with the routing
67 model attribute and eventually with some \<route\> tag). You define AS
68 by using ... well ... the \<AS\> tag. An AS can also contain some AS :
69 AS allows you to define the hierarchy of your platform.
71 Within each AS, you basically have the following type of resources:
72 \li <b>host</b>: an host, with cores in it, and so on
73 \li <b>router</b>: a router or a gateway.
74 \li <b>link</b>: a link, that defines a connection between two (or
75 more) resources (and have a bandwidth and a latency)
76 \li <b>cluster</b>: like a real cluster, contains many hosts
77 interconnected by some dedicated network.
79 Between those elements, a routing has to be defined. As the AS is
80 supposed to be Autonomous, this has to be done at the AS level. As AS
81 handles two different types of entities (<b>host/router</b> and
82 <b>AS</b>) you will have to define routes between those elements. A
83 network model have to be provided for AS, but you may/will need,
84 depending of the network model, or because you want to bypass the
85 default behavior to defines routes manually. There are 3 tags to use:
86 \li <b>ASroute</b>: to define routes between two <b>AS</b>
87 \li <b>route</b>: to define routes between two <b>host/router</b>
88 \li <b>bypassRoute</b>: to define routes between two <b>AS</b> that
89 will bypass default routing.
91 Here is an illustration of the overall concepts:
94 <a href="AS_hierarchy.png" border=0><img src="AS_hierarchy.png" width="30%" border=0 align="center"></a>
97 Circles represent processing units and squares represent network routers. Bold
98 lines represent communication links. AS2 models the core of a national
99 network interconnecting a small flat cluster (AS4) and a larger
100 hierarchical cluster (AS5), a subset of a LAN (AS6), and a set of peers
101 scattered around the world (AS7).
104 This is all for the concepts ! To make a long story short, a SimGrid
105 platform is made of a hierarchy of AS, each of them containing
106 resources, and routing is defined at AS level. Let's have a deeper
111 \section pf_pftags Describing resources and their organization
113 \subsection pf_As Platform organization tag : AS
115 AS (or Autonomous System) is an organizational unit that contains
116 resources and defines routing between them, and eventually some other
117 AS. So it allows you to define a hierarchy into your platform.
118 <b>*ANY*</b> resource <b>*MUST*</b> belong to an AS. There are a few
121 <b>AS</b> attributes :
122 \li <b>id (mandatory)</b>: the identifier of AS to be used when
124 \li <b>routing (mandatory)</b>: the routing model used into it. By
125 model we mean the internal way the simulator will manage routing.
126 That also have a big impact on how many information you'll have to
127 provide to help the simulator to route between the AS elements.
128 <b>routing</b> possible values are <b>Full, Floyd, Dijkstra,
129 DijkstraCache, none, Vivaldi, Cluster</b>. For more
130 explanation about what to choose, take a look at the section
133 Elements into an AS are basically resources (computers, network
134 equipments) and some routing information if necessary (see below for
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.
151 \subsection pf_Cr Computing resources: hosts, clusters and peers.
153 \subsubsection pf_host host
155 A <b>host</b> represents a computer, where you will be able to execute
156 code and from which you can send and receive information. A host can
157 contain more than 1 core. Here are the attributes of a host :
160 <b>host</b> attributes :
161 \li <b>id (mandatory)</b>: the identifier of the host to be used when
163 \li <b>power (mandatory)</b>:the peak number FLOPS the CPU can manage.
165 \li <b>core</b>: The number of core of this host (by default, 1). If
166 you specify the amount of cores, the 'power' parameter is the power
168 For example, if you specify that your host has 6 cores, it will be
169 available to up to 6 sequential tasks without sharing. If more
170 tasks are placed on this host, the resource will be shared
171 accordingly. For example, if you schedule 12 tasks on that host,
172 each will get half of the specified computing power. Please note
173 that although sound, this model were never scientifically assessed.
174 Please keep this fact in mind when using it.
175 \li <b>availability</b>: specify if the percentage of power available.
176 \li <b>availability_file</b>: Allow you to use a file as input. This
177 file will contain availability traces for this computer. The
178 syntax of this file is defined below. Possible values : absolute
179 or relative path, syntax similar to the one in use on your system.
180 \li <b>state</b>: the computer state, as in : is that computer ON or
181 OFF. Possible values : "ON" or "OFF".
182 \li <b>state_file</b>: Same mechanism as availability_file, similar
184 \li <b>coordinates</b>: you'll have to give it if you choose the
185 vivaldi, coordinate-based routing model for the AS the host
186 belongs to. More details about it in the P2P coordinate based
189 An host can contain some <b>mount</b> that defines mounting points
190 between some storage resource and the <b>host</b>. Please refer to the
191 storage doc for more information.
193 An host can also contain the <b>prop</b> tag. the prop tag allows you
194 to define additional information on this host following the
195 attribute/value schema. You may want to use it to give information to
196 the tool you use for rendering your simulation, for example.
200 <host id="host1" power="1000000000"/>
201 <host id="host2" power="1000000000">
202 <prop id="color" value="blue"/>
203 <prop id="rendershape" value="square"/>
208 <b>Expressing dynamicity.</b>
209 It is also possible to seamlessly declare a host whose
210 availability changes over time using the availability_file
211 attribute and a separate text file whose syntax is exemplified below.
213 <b>Adding a trace file</b>
215 <platform version="1">
216 <host id="bob" power="500000000"
217 availability_file="bob.trace" />
220 <b>Example of "bob.trace" file</b>
228 At time 0, our host will deliver 500~Mflop/s. At time 11.0, it will
229 deliver half, that is 250~Mflop/s until time 20.0 where it will
230 will start delivering 80\% of its power, that is 400~Mflop/s. Last, at
231 time 21.0 (20.0 plus the periodicity 1.0), we loop back to the
232 beginning and the host will deliver again 500~Mflop/s.
234 <b>Changing initial state</b>
236 It is also possible to specify whether the host
237 is up or down by setting the <b>state</b> attribute to either <b>ON</b>
238 (default value) or <b>OFF</b>.
240 <b>Expliciting the default value "ON"</b>
242 <platform version="1">
248 <b>Host switched off</b>
250 <platform version="1">
256 <b>Expressing churn</b>
257 To express the fact that a host can change state over time (as in P2P
258 systems, for instance), it is possible to use a file describing the time
259 at which the host is turned on or off. An example of the content
260 of such a file is presented below.
261 <b>Adding a state file</b>
263 <platform version="1">
264 <host id="bob" power="500000000"
265 state_file="bob.fail" />
268 <b>Example of "bob.fail" file</b>
275 A negative value means <b>down</b> while a positive one means <b>up and
276 running</b>. From time 0.0 to time 1.0, the host is on. At time 1.0, it is
277 turned off and at time 2.0, it is turned on again until time 12 (2.0 plus the
278 periodicity 10.0). It will be turned on again at time 13.0 until time 23.0, and
283 \subsubsection pf_cluster cluster
285 A <b>cluster</b> represents a cluster. It is most of the time used
286 when you want to have a bunch of machine defined quickly. It must be
287 noted that cluster is meta-tag : <b>from the inner SimGrid point of
288 view, a cluster is an AS where some optimized routing is defined</b>.
289 The default inner organization of the cluster is as follow:
295 ____________|__________|_____________ backbone
297 l0| l1| l2| l97| l96 | | l99
303 You have a set of <b>host</b> defined. Each of them has a <b>link</b>
304 to a central backbone (backbone is a <b>link</b> itself, as a link can
305 be used to represent a switch, see the switch or <b>link</b> section
306 below for more details about it). A <b>router</b> gives a way to the
307 <b>cluster</b> to be connected to the outside world. Internally,
308 cluster is then an AS containing all hosts : the router is the default
309 gateway for the cluster.
311 There is an alternative organization, which is as follow :
325 The principle is the same, except we don't have the backbone. The way
326 to obtain it is simple : you just have to let bb_* attributes
331 <b>cluster</b> attributes :
332 \li <b>id (mandatory)</b>: the identifier of the cluster to be used
333 when referring to it.
334 \li <b>prefix (mandatory)</b>: each node of the cluster has to have a
335 name. This is its prefix.
336 \li <b>suffix (mandatory)</b>: node suffix name.
337 \li <b>radical (mandatory)</b>: regexp used to generate cluster nodes
338 name. Syntax is quite common, "10-20" will give you 11 machines
339 numbered from 10 to 20, "10-20;2" will give you 12 machines, one
340 with the number 2, others numbered as before. The produced number
341 is concatenated between prefix and suffix to form machine names.
342 \li <b>power (mandatory)</b>: same as <b>host</b> power.
343 \li <b>core</b>: same as <b>host</b> core.
344 \li <b>bw (mandatory)</b>: bandwidth for the links between nodes and
345 backbone (if any). See <b>link</b> section for syntax/details.
346 \li <b>lat (mandatory)</b>: latency for the links between nodes and
347 backbone (if any). See <b>link</b> section for syntax/details.
348 \li <b>sharing_policy</b>: sharing policy for the links between nodes
349 and backbone (if any). See <b>link</b> section for syntax/details.
350 \li <b>bb_bw </b>: bandwidth for backbone (if any). See <b>link</b>
351 section for syntax/details. If both bb_* attributes are omitted,
352 no backbone is created (alternative cluster architecture described
354 \li <b>bb_lat </b>: latency for backbone (if any). See <b>link</b>
355 section for syntax/details. If both bb_* attributes are omitted,
356 no backbone is created (alternative cluster architecture described
358 \li <b>bb_sharing_policy</b>: sharing policy for the backbone (if
359 any). See <b>link</b> section for syntax/details.
360 \li <b>availability_file</b>: Allow you to use a file as input for
361 availability. Similar to <b>hosts</b> attribute.
362 \li <b>state_file</b>: Allow you to use a file as input for states.
363 Similar to <b>hosts</b> attribute.
365 the router name is defined as the resulting String in the following
369 router_name = prefix + clusterId + _router + suffix;
373 <b>cluster example</b>
375 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
376 power="1e9" bw="125e6" lat="5E-5"/>
378 <cluster id="my_cluster_1" prefix="c-" suffix=".me" radical="0-99"
379 power="1e9" bw="125e6" lat="5E-5"
380 bb_bw="2.25e9" bb_lat="5E-4"/>
382 The second examples creates one router and 100 machines, which names
385 c-my_cluster_1_router.me
393 \subsubsection pf_peer peer
394 A <b>peer</b> represents a peer, as in Peer-to-Peer (P2P). Basically,
395 as cluster, <b>A PEER IS INTERNALLY INTERPRETED AS AN \<AS\></b>. It's
396 just a kind of shortcut that does the following :
398 \li It creates a tiny AS whose routing type is cluster
399 \li It creates an host
400 \li Two links : one for download and one for upload. This is
401 convenient to use and simulate stuff under the last mile model (as
403 \li It connects the two links to the host
404 \li It creates a router (a gateway) that serve as entry point for this peer zone.
405 This router has coordinates.
407 <b>peer</b> attributes :
408 \li <b>id (mandatory)</b>: the identifier of the peer to be used when
410 \li <b>power CDATA (mandatory)</b>: as in host
411 \li <b>bw_in CDATA (mandatory)</b>: bandwidth in.
412 \li <b>bw_out CDATA (mandatory)</b>:bandwidth out.
413 \li <b>lat CDATA (mandatory)</b>: Latency for in and out links.
414 \li <b>coordinates</b>: coordinates of the gateway for this peer.
415 \li <b>sharing_policy</b>: sharing policy for links. Can be SHARED or
416 FULLDUPLEX, FULLDUPLEX is the default. See <b>link</b> description
418 \li <b>availability_file</b>: availability file for the peer. Same as
419 host availability file. See <b>host</b> description for details.
420 \li <b>state_file </b>: state file for the peer. Same as host state
421 file. See <b>host</b> description for details.
423 In term of XML, the <b>peer</b> construct can be explained as follows: it transforms
426 coordinates="12.8 14.4 6.4"
434 <AS id="as_FOO" routing="Cluster">
435 <host id="peer_FOO" power="1.5Gf"/>
436 <link id="link_FOO_UP" bandwidth="2.25GBps" latency="500us"/>
437 <link id="link_FOO_DOWN" bandwidth="2.25GBps" latency="500us"/>
438 <router id="router_FOO" coordinates="25.5 9.4 1.4"/>
439 <host_link id="peer_FOO" up="link_FOO_UP" down="link_FOO_DOWN"/>
444 \subsection pf_ne Network equipments: links and routers
446 You have basically two entities available to represent network entities:
447 \li <b>link</b>: represents something that has a limited bandwidth, a
448 latency, and that can be shared according to TCP way to share this
449 bandwidth. <b>LINKS ARE NOT EDGES BUT HYPEREDGES</b>: it means
450 that you can have more than 2 equipments connected to it.
451 \li <b>router</b>: represents something that one message can be routed
452 to, but does not accept any code, nor have any influence on the
453 performances (no bandwidth, no latency, not anything).<b>ROUTERS
454 ARE ENTITIES (ALMOST) IGNORED BY THE SIMULATOR WHEN THE SIMULATION
455 HAS BEGUN</b>. If you want to represent something like a switch,
456 you must use <b>link</b> (see section below). Routers are used in
457 order to run some routing algorithm and determine routes (see
458 routing section for details).
460 let's see deeper what those entities hide.
462 \subsubsection pf_router router
463 As said before, <b>router</b> is used only to give some information
464 for routing algorithms. So, it does not have any attributes except :
466 <b>router</b> attributes :
467 \li <b>id (mandatory)</b>: the identifier of the router to be used
468 when referring to it.
469 \li <b>coordinates</b>: you'll have to give it if you choose the
470 vivaldi, coordinate-based routing model for the AS the host
471 belongs to. More details about it in the P2P coordinates based
474 <b>router example</b>
476 <router id="gw_dc1_horizdist"/>
479 \subsubsection pf_link link
481 Network links can represent one-hop network connections. They are
482 characterized by their id and their bandwidth. The latency is optional
483 with a default value of 0.0. For instance, we can declare a network
484 link named link1 having bandwidth of 1Gb/s and a latency of 50µs.
488 <link id="LINK1" bandwidth="125000000" latency="5E-5"/>
490 <b>Expressing sharing policy</b>
492 By default a network link is SHARED, that is if more than one flow go
493 through a link, each gets a share of the available bandwidth similar
494 to the share TCP connections offers.
496 Conversely if a link is defined as a FATPIPE, each flow going through
497 this link will get all the available bandwidth, whatever the number of
498 flows. The FATPIPE behavior allows to describe big backbones that
499 won't affect performances (except latency). Finally a link can be
500 considered as FULLDUPLEX, that means that in the simulator, 2 links
501 (one named UP and the other DOWN) will be created for each link, so as
502 the transfers from one side to the other will interact similarly as
503 TCP when ACK returning packets circulate on the other direction. More
504 discussion about it is available in <b>link_ctn</b> description.
507 <link id="SWITCH" bandwidth="125000000" latency="5E-5" sharing_policy="FATPIPE" />
510 <b>Expressing dynamicity and failures</b>
512 As for hosts, it is possible to declare links whose state, bandwidth
513 or latency change over the time. In this case, the bandwidth and
514 latency attributes are respectively replaced by the bandwidth file and
515 latency file attributes and the corresponding text files.
518 <link id="LINK1" state_file="link1.fail" bandwidth="80000000" latency=".0001" bandwidth_file="link1.bw" latency_file="link1.lat" />
521 It has to be noted that even if the syntax is the same, the semantic
522 of bandwidth and latency trace files differs from that of host
523 availability files. Those files do not express availability as a
524 fraction of the available capacity but directly in bytes per seconds
525 for the bandwidth and in seconds for the latency. This is because most
526 tools allowing to capture traces on real platforms (such as NWS)
527 express their results this way.
529 <b>Example of "link1.bw" file</b>
536 <b>Example of "link1.lat" file</b>
544 In this example, the bandwidth varies with a period of 12 seconds
545 while the latency varies with a period of 5 seconds. At the beginning
546 of simulation, the link’s bandwidth is of 80,000,000 B/s (i.e., 80
547 Mb/s). After four seconds, it drops at 40 Mb/s, and climbs back to 60
548 Mb/s after eight seconds. It keeps that way until second 12 (ie, until
549 the end of the period), point at which it loops its behavior (seconds
550 12-16 will experience 80 Mb/s, 16-20 40 Mb/s and so on). In the same
551 time, the latency values are 100µs (initial value) on the [0, 1[ time
552 interval, 1ms on [1, 2[, 10ms on [2, 3[, 1ms on [3,5[ (i.e., until the
553 end of period). It then loops back, starting at 100µs for one second.
555 <b>link</b> attributes :
556 \li <b>id (mandatory)</b>: the identifier of the link to be used when referring to it.
557 \li <b>bandwidth (mandatory)</b>: bandwidth for the link.
558 \li <b>lat </b>: latency for the link. Default is 0.0.
559 \li <b>sharing_policy</b>: sharing policy for the link.
560 \li <b>state</b>: Allow you to to set link as ON or OFF. Default is ON.
561 \li <b>bandwidth_file</b>: Allow you to use a file as input for bandwidth.
562 \li <b>latency_file</b>: Allow you to use a file as input for latency.
563 \li <b>state_file</b>: Allow you to use a file as input for states.
565 As an host, a <b>link</b> tag can also contain the <b>prop</b> tag.
569 <link id="link1" bandwidth="125000000" latency="0.000100"/>
573 \subsection pf_storage Storage
575 <b>Note : This is a prototype version that should evolve quickly, this
576 is just some doc valuable only at the time of writing this doc</b>
577 This section describes the storage management under SimGrid ; nowadays
578 it's only usable with MSG. It relies basically on linux-like concepts.
579 You also may want to have a look to its corresponding section in \ref
580 msg_file_management ; functions access are organized as a POSIX-like
583 \subsubsection pf_sto_conc Storage Main concepts
584 Basically there is 3 different entities to know :
585 \li the <b>storage_type</b>: here you define some kind of storage that
586 you will instantiate many type on your platform. Think of it like
587 a definition of throughput of a specific disk.
588 \li the <b>storage</b>: instance of a <b>storage_type</b>. Defines a
589 new storage of <b>storage_type</b>
590 \li the <b>mount</b>: says that the storage is located into this
593 the content of a storage has to be defined in a content file that
594 contains the content. The path to this file has to be passed within
595 the <b>content</b> attribute . Here is a way to generate it:
598 find /path/you/want -type f -exec ls -l {} \; 2>/dev/null > ./content.txt
601 \subsubsection pf_sto_sttp storage_type
604 <b>storage_type</b> attributes :
605 \li <b>id (mandatory)</b>: the identifier of the storage_type to be
606 used when referring to it.
607 \li <b>model (mandatory)</b>: Unused for now by the simulator (but
609 \li <b>content</b>: default value 0. The file containing the disk
610 content. (may be moved soon or later to <b>storage</b> tag.
612 The tag must contains some predefined model prop, as may do some other
614 <b>storage_type</b> mandatory <b>model_prop</b> :
615 \li <b>Bwrite</b>: value in B/s. Write throughput
616 \li <b>Bread</b>: value in B/s. Read throughput
617 \li <b>Bconnexion</b>: value in B/s. Connection throughput (i.e. the
618 throughput of the storage connector).
620 A storage_type can also contain the <b>prop</b> tag. The prop tag allows you
621 to define additional information on this storage_type following the
622 attribute/value schema. You may want to use it to give information to
623 the tool you use for rendering your simulation, for example.
626 <storage_type id="single_HDD" model="linear_no_lat" size="4000" content_type="txt_unix">
627 <model_prop id="Bwrite" value="30MBps" />
628 <model_prop id="Bread" value="100MBps" />
629 <model_prop id="Bconnection" value="150MBps" />
630 <b><prop id="Brand" value="Western Digital" /></b>
634 \subsubsection pf_sto_st storage
636 <b>storage_type</b> attributes :
637 \li <b>id (mandatory)</b>: the identifier of the storage to be used
638 when referring to it.
639 \li <b>typeId (mandatory)</b>: the identifier of the storage_type that
640 this storage belongs to.
641 \li <b>attach (mandatory)</b>: the host (name) to which the storage is
644 \subsubsection pf_sto_mo mount
646 <b>mount</b> attributes :
647 \li <b>id (mandatory)</b>: the id of the <b>storage</b> that must be
648 mounted on that computer.
649 \li <b>name (mandatory)</b>: the name that will be the logical
650 reference to this disk (the mount point).
652 \subsubsection pf_sto_mst mstorage
653 <b>Note : unused for now</b>
654 <b>mstorage</b> attributes :
655 \li <b>typeId (mandatory)</b>: the id of the <b>storage</b> that must
656 be mounted on that computer.
657 \li <b>name (mandatory)</b>: the name that will be the logical
658 reference to this disk (the mount point).
660 \section pf_routing Routing
662 In order to run fast, it has been chosen to use static routing within
663 SimGrid. By static, it means that it is calculated once (or almost),
664 and will not change during execution. We chose to do that because it
665 is rare to have a real deficiency of a resource ; most of the time, a
666 communication fails because the links are too overloaded, and so your
667 connection stops before the time out, or because the computer at the
668 other end is not answering.
670 We also chose to use shortest paths algorithms in order to emulate
671 routing. Doing so is consistent with the reality: RIP, OSPF, BGP are
672 all calculating shortest paths. They have some convergence time, but
673 at the end, so when the platform is stable (and this should be the
674 moment you want to simulate something using SimGrid) your packets will
675 follow the shortest paths.
677 \subsection pf_rm Routing models
679 Within each AS, you have to define a routing model to use. You have
680 basically 3 main kind of routing models :
682 \li Shortest-path based models: you let SimGrid calculates shortest
683 paths and manage it. Behaves more or less as most real life
685 \li Manually-entered route models: you'll have to define all routes
686 manually by yourself into the platform description file.
687 Consistent with some manually managed real life routing.
688 \li Simple/fast models: those models offers fast, low memory routing
689 algorithms. You should consider to use it if you can make some
690 assumptions about your AS. Routing in this case is more or less
693 \subsubsection pf_raf The router affair
695 Expressing routers becomes mandatory when using shortest-path based
696 models or when using ns-3 or the bindings to the GTNetS packet-level
697 simulator instead of the native analytical network model implemented
700 For graph-based shortest path algorithms, routers are mandatory,
701 because both algorithms need a graph, and so we need to have source
702 and destination for each edge.
704 Routers are naturally an important concept in GTNetS or ns-3 since the
705 way they run the packet routing algorithms is actually simulated.
706 Instead, the SimGrid’s analytical models aggregate the routing time
707 with the transfer time. Rebuilding a graph representation only from
708 the route information turns to be a very difficult task, because of
709 the missing information about how routes intersect. That is why we
710 introduced a \<router\> tag, which is simply used to express these
711 intersection points. The only attribute accepted by this tag an id. It
712 is important to understand that the \<router\> tag is only used to
713 provide topological information.
715 To express those topological information, some <b>route</b> have to be
716 defined saying which link is between which routers. Description or the
717 route syntax is given below, as well as example for the different
720 \subsubsection pf_rm_sh Shortest-path based models
722 Here is the complete list of such models, that computes routes using
723 classic shortest-paths algorithms. How to choose the best suited
724 algorithm is discussed later in the section devoted to it.
726 \li <b>Floyd</b>: Floyd routing data. Pre-calculates all routes once.
727 \li <b>Dijkstra</b>: Dijkstra routing data ,calculating routes when
729 \li <b>DijkstraCache</b>: Dijkstra routing data. Handle some cache for
730 already calculated routes.
732 All those shortest-path models are instanciated the same way. Here are
737 <AS id="AS0" routing="Floyd">
739 <cluster id="my_cluster_1" prefix="c-" suffix=""
740 radical="0-1" power="1000000000" bw="125000000" lat="5E-5"
741 router_id="router1"/>
743 <AS id="AS1" routing="none">
744 <host id="host1" power="1000000000"/>
747 <link id="link1" bandwidth="100000" latency="0.01"/>
749 <ASroute src="my_cluster_1" dst="AS1"
752 <link_ctn id="link1"/>
758 ASroute given at the end gives a topological information: link1 is
759 between router1 and host1.
763 <AS id="AS_2" routing="Dijsktra">
764 <host id="AS_2_host1" power="1000000000"/>
765 <host id="AS_2_host2" power="1000000000"/>
766 <host id="AS_2_host3" power="1000000000"/>
767 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
768 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
769 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
770 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
771 <router id="central_router"/>
772 <router id="AS_2_gateway"/>
773 <!-- routes providing topological information -->
774 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
775 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
776 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
777 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
781 DijsktraCache example :
783 <AS id="AS_2" routing="DijsktraCache">
784 <host id="AS_2_host1" power="1000000000"/>
786 (platform unchanged compared to upper example)
789 \subsubsection pf_rm_me Manually-entered route models
791 \li <b>Full</b>: You have to enter all necessary routes manually
795 <AS id="AS0" routing="Full">
796 <host id="host1" power="1000000000"/>
797 <host id="host2" power="1000000000"/>
798 <link id="link1" bandwidth="125000000" latency="0.000100"/>
799 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
803 \subsubsection pf_rm_sf Simple/fast models
805 \li <b>none</b>: No routing (Unless you know what you are doing, avoid
806 using this mode in combination with a non Constant network model).
809 <AS id="exitAS" routing="none">
810 <router id="exit_gateway"/>
813 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use
814 coordinates. See the corresponding section P2P below for details.
815 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
816 not be used, except internally.
818 \subsection ps_dec Defining routes
820 The principle of route definition is the same for the 4 available tags
821 for doing it. Those for tags are:
823 \li <b>route</b>: to define route between host/router
824 \li <b>ASroute</b>: to define route between AS
825 \li <b>bypassRoute</b>: to bypass normal routes as calculated by the
826 network model between host/router
827 \li <b>bypassASroute</b>: same as bypassRoute, but for AS
829 Basically all those tags will contain an (ordered) list of references
830 to link that compose the route you want to define.
832 Consider the example below:
835 <route src="Alice" dst="Bob">
836 <link_ctn id="link1"/>
837 <link_ctn id="link2"/>
838 <link_ctn id="link3"/>
842 The route here from host Alice to Bob will be first link1, then link2,
843 and finally link3. What about the reverse route ? <b>route</b> and
844 <b>ASroute</b> have an optional attribute <b>symmetrical</b>, that can
845 be either YES or NO. YES means that the reverse route is the same
846 route in the inverse order, and is set to YES by default. Note that
847 this is not the case for bypass*Route, as it is more probable that you
848 want to bypass only one default route.
850 For an ASroute, things are just slightly more complicated, as you have
851 to give the id of the gateway which is inside the AS you're talking
852 about you want to access ... So it looks like this :
856 <ASroute src="AS1" dst="AS2"
857 gw_src="router1" gw_dst="router2">
858 <link_ctn id="link1"/>
862 gw == gateway, so when any message are trying to go from AS1 to AS2,
863 it means that it must pass through router1 to get out of the AS, then
864 pass through link1, and get into AS2 by being received by router2.
865 router1 must belong to AS1 and router2 must belong to AS2.
867 \subsubsection pf_linkctn link_ctn
869 a <b>link_ctn</b> is the tag that is used in order to reference a
870 <b>link</b> in a route. Its id is the link id it refers to.
872 <b>link_ctn</b> attributes :
873 \li <b>id (mandatory)</b>: Id of the link this tag refers to
874 \li <b>direction</b>: if the link referenced by <b>id</b> has been
875 declared as FULLDUPLEX, this is used to indicate in which
876 direction the route you're defining is going through this link.
877 Possible values "UP" or "DOWN".
879 \subsubsection pf_asro ASroute
881 ASroute tag purpose is to let people write manually their routes
882 between AS. It's useful when you're in Full model.
884 <b>ASroute</b> attributes :
885 \li <b>src (mandatory)</b>: the source AS id.
886 \li <b>dst (mandatory)</b>: the destination AS id.
887 \li <b>gw_src (mandatory)</b>: the gateway to be used within the AS.
888 Can be any <b>host</b> or \b router defined into the \b src AS or
889 into one of the AS it includes.
890 \li <b>gw_dst (mandatory)</b>: the gateway to be used within the AS.
891 Can be any <b>host</b> or \b router defined into the \b dst AS or
892 into one of the AS it includes.
893 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
894 will be the opposite of the one defined. Can be either YES or NO,
897 <b>Example of ASroute with Full</b>
899 <AS id="AS0" routing="Full">
900 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
901 radical="0-149" power="1000000000" bw="125000000" lat="5E-5"
902 bb_bw="2250000000" bb_lat="5E-4"/>
904 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
905 radical="150-299" power="1000000000" bw="125000000" lat="5E-5"
906 bb_bw="2250000000" bb_lat="5E-4"/>
908 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
910 <ASroute src="my_cluster_1" dst="my_cluster_2"
911 gw_src="c-my_cluster_1_router.me"
912 gw_dst="c-my_cluster_2_router.me">
913 <link_ctn id="backbone"/>
915 <ASroute src="my_cluster_2" dst="my_cluster_1"
916 gw_src="c-my_cluster_2_router.me"
917 gw_dst="c-my_cluster_1_router.me">
918 <link_ctn id="backbone"/>
923 \subsubsection pf_ro route
924 The principle is the same as ASroute : <b>route</b> contains list of
925 links that are in the path between src and dst, except that it is for
926 routes between a src that can be either <b>host</b> or \b router and a
927 dst that can be either <b>host</b> or \b router. Useful for Full
928 as well as for the shortest-paths based models, where you
929 have to give topological information.
932 <b>route</b> attributes :
933 \li <b>src (mandatory)</b>: the source id.
934 \li <b>dst (mandatory)</b>: the destination id.
935 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
936 will be the opposite of the one defined. Can be either YES or NO,
939 <b>route example in Full</b>
941 <route src="Tremblay" dst="Bourassa">
942 <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"/>
946 <b>route example in a shortest-path model</b>
948 <route src="Tremblay" dst="Bourassa">
952 Note that when using route to give topological information, you have
953 to give routes with one link only in it, as SimGrid needs to know
954 which host are at the end of the link.
956 \subsubsection pf_byro bypassASroute
958 <b>Note : bypassASroute and bypassRoute are under rewriting to perform
959 better ; so you may not use it yet</b> As said before, once you choose
960 a model, it (if so) calculates routes for you. But maybe you want to
961 define some of your routes, which will be specific. You may also want
962 to bypass some routes defined in lower level AS at an upper stage :
963 <b>bypassASroute</b> is the tag you're looking for. It allows to
964 bypass routes defined between already defined between AS (if you want
965 to bypass route for a specific host, you should just use byPassRoute).
966 The principle is the same as ASroute : <b>bypassASroute</b> contains
967 list of links that are in the path between src and dst.
969 <b>bypassASroute</b> attributes :
970 \li <b>src (mandatory)</b>: the source AS id.
971 \li <b>dst (mandatory)</b>: the destination AS id.
972 \li <b>gw_src (mandatory)</b>: the gateway to be used within the AS.
973 Can be any <b>host</b> or \b router defined into the \b src AS or
974 into one of the AS it includes.
975 \li <b>gw_dst (mandatory)</b>: the gateway to be used within the AS.
976 Can be any <b>host</b> or \b router defined into the \b dst AS or
977 into one of the AS it includes.
978 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
979 will be the opposite of the one defined. Can be either YES or NO,
982 <b>bypassASroute Example</b>
984 <bypassASRoute src="my_cluster_1" dst="my_cluster_2"
985 gw_src="my_cluster_1_router"
986 gw_dst="my_cluster_2_router">
987 <link_ctn id="link_tmp"/>
991 \subsubsection pf_byro bypassRoute
992 <b>Note : bypassASRoute and bypassRoute are under rewriting to perform
993 better ; so you may not use it yet</b> As said before, once you choose
994 a model, it (if so) calculates routes for you. But maybe you want to
995 define some of your routes, which will be specific. You may also want
996 to bypass some routes defined in lower level AS at an upper stage :
997 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
998 routes defined between <b>host/router</b>. The principle is the same
999 as route : <b>bypassRoute</b> contains list of links references of
1000 links that are in the path between src and dst.
1002 <b>bypassRoute</b> attributes :
1003 \li <b>src (mandatory)</b>: the source AS id.
1004 \li <b>dst (mandatory)</b>: the destination AS id.
1005 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
1006 will be the opposite of the one defined. Can be either YES or NO,
1009 <b>bypassRoute Example</b>
1011 <b>bypassRoute Example</b>
1013 <bypassRoute src="host_1" dst="host_2">
1014 <link_ctn id="link_tmp"/>
1019 \subsection pb_baroex Basic Routing Example
1021 Let's say you have an AS named AS_Big that contains two other AS, AS_1
1022 and AS_2. If you want to make an host (h1) from AS_1 with another one
1023 (h2) from AS_2 then you'll have to proceed as follow:
1024 \li First, you have to ensure that a route is defined from h1 to the
1025 AS_1's exit gateway and from h2 to AS_2's exit gateway.
1026 \li Then, you'll have to define a route between AS_1 to AS_2. As those
1027 AS are both resources belonging to AS_Big, then it has to be done
1028 at AS_big level. To define such a route, you have to give the
1029 source AS (AS_1), the destination AS (AS_2), and their respective
1030 gateway (as the route is effectively defined between those two
1031 entry/exit points). Elements of this route can only be elements
1032 belonging to AS_Big, so links and routers in this route should be
1033 defined inside AS_Big. If you choose some shortest-path model,
1034 this route will be computed automatically.
1036 As said before, there are mainly 2 tags for routing :
1037 \li <b>ASroute</b>: to define routes between two <b>AS</b>
1038 \li <b>route</b>: to define routes between two <b>host/router</b>
1040 As we are dealing with routes between AS, it means that those we'll
1041 have some definition at AS_Big level. Let consider AS_1 contains 1
1042 host, 1 link and one router and AS_2 3 hosts, 4 links and one router.
1043 There will be a central router, and a cross-like topology. At the end
1044 of the crosses arms, you'll find the 3 hosts and the router that will
1045 act as a gateway. We have to define routes inside those two AS. Let
1046 say that AS_1 contains full routes, and AS_2 contains some Floyd
1047 routing (as we don't want to bother with defining all routes). As
1048 we're using some shortest path algorithms to route into AS_2, we'll
1049 then have to define some <b>route</b> to gives some topological
1050 information to SimGrid. Here is a file doing it all :
1053 <AS id="AS_Big" routing="Dijsktra">
1054 <AS id="AS_1" routing="Full">
1055 <host id="AS_1_host1" power="1000000000"/>
1056 <link id="AS_1_link" bandwidth="1250000000" latency="5E-4"/>
1057 <router id="AS_1_gateway"/>
1058 <route src="AS_1_host1" dst="AS_1_gateway">
1059 <link_ctn id="AS_1_link"/>
1062 <AS id="AS_2" routing="Floyd">
1063 <host id="AS_2_host1" power="1000000000"/>
1064 <host id="AS_2_host2" power="1000000000"/>
1065 <host id="AS_2_host3" power="1000000000"/>
1066 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1067 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1068 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1069 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1070 <router id="central_router"/>
1071 <router id="AS_2_gateway"/>
1072 <!-- routes providing topological information -->
1073 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1074 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1075 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1076 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1078 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1080 <ASroute src="AS_1" dst="AS_2"
1081 gw_src="AS_1_gateway"
1082 gw_dst="AS_2_gateway">
1083 <link_ctn id="backbone"/>
1088 \section pf_other_tags Tags not (directly) describing the platform
1090 There are 3 tags, that you can use inside a \<platform\> tag that are
1091 not describing the platform:
1092 \li random: it allows you to define random generators you want to use
1093 for your simulation.
1094 \li config: it allows you to pass some configuration stuff like, for
1095 example, the network model and so on. It follows the
1096 \li include: simply allows you to include another file into the
1099 \subsection pf_conf config
1100 <b>config</b> attributes :
1101 \li <b>id (mandatory)</b>: the identifier of the config to be used
1102 when referring to it.
1105 <b>config</b> tag only purpose is to include <b>prop</b> tags. Valid
1106 id are basically the same as the list of possible parameters you can
1107 use by command line, except that "/" are used for namespace
1108 definition. See the \ref options config and options page for more
1112 <b>config example</b>
1114 <?xml version='1.0'?>
1115 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1116 <platform version="3">
1117 <config id="General">
1118 <prop id="maxmin/precision" value="0.000010"></prop>
1119 <prop id="cpu/optim" value="TI"></prop>
1120 <prop id="workstation/model" value="compound"></prop>
1121 <prop id="network/model" value="SMPI"></prop>
1122 <prop id="path" value="~/"></prop>
1123 <prop id="smpi/bw_factor" value="65472:0.940694;15424:0.697866;9376:0.58729"></prop>
1126 <AS id="AS0" routing="Full">
1131 \subsection pf_rand random
1132 Not yet in use, and possibly subject to huge modifications.
1134 \subsection pf_incl include
1135 <b>include</b> tag allows to import into a file platform parts located
1136 in another file. This is done with the intention to help people
1137 combine their different AS and provide new platforms. Those files
1138 should contains XML part that contains either
1139 <b>include,cluster,peer,AS,trace,trace_connect</b> tags.
1141 <b>include</b> attributes :
1142 \li <b>file (mandatory)</b>: filename of the file to include. Possible
1143 values: absolute or relative path, syntax similar to the one in
1146 <b>Note</b>: due to some obscure technical reasons, you have to open
1147 and close tag in order to let it work.
1148 <b>include Example</b>
1150 <?xml version='1.0'?>
1151 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1152 <platform version="3">
1153 <AS id="main" routing="Full">
1154 <include file="clusterA.xml"></include>
1155 <include file="clusterB.xml"></include>
1160 \subsection pf_tra trace and trace_connect
1161 Both tags are an alternate way to passe availability, state, and so on
1162 files to entity. Instead of referring to the file directly in the host,
1163 link, or cluster tag, you proceed by defining a trace with an id
1164 corresponding to a file, later an host/link/cluster, and finally using
1165 trace_connect you say that the file trace must be used by the entity.
1166 Get it ? Let's have a look at an example :
1169 <AS id="AS0" routing="Full">
1170 <host id="bob" power="1000000000"/>
1172 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1173 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1176 All constraints you have is that <b>trace_connect</b> is after
1177 <b>trace</b> and <b>host</b> definitions.
1180 <b>trace</b> attributes :
1181 \li <b>id (mandatory)</b>: the identifier of the trace to be used when
1183 \li <b>file</b>: filename of the file to include. Possible values :
1184 absolute or relative path, syntax similar to the one in use on
1185 your system. If omitted, the system expects that you provide the
1186 trace values inside the trace tags (see below).
1187 \li <b>trace periodicity (mandatory)</b>: trace periodicity, same
1188 definition as in hosts (see upper for details).
1190 Here is an example of trace when no file name is provided:
1193 <trace id="myTrace" periodicity="1.0">
1200 <b>trace_connect</b> attributes :
1201 \li <b>kind</b>: the type of trace, possible values
1202 <b>HOST_AVAIL|POWER|LINK_AVAIL|BANDWIDTH|LATENCY,</b> default:
1204 \li <b>trace (mandatory)</b>: the identifier of the trace referenced.
1205 \li <b>element (mandatory)</b>: the identifier of the entity referenced.
1209 \section pf_hints Hints and tips, or how to write a platform efficiently
1211 Now you should know at least the syntax and be able to create a
1212 platform by your own. However, after having ourselves wrote some platforms, there
1213 are some best practices you should pay attention to in order to
1214 produce good platform and some choices you can make in order to have
1215 faster simulations. Here's some hints and tips, then.
1217 \subsection pf_as_h AS Hierarchy
1218 The AS design allows SimGrid to go fast, because computing route is
1219 done only for the set of resources defined in this AS. If you're using
1220 only a big AS containing all resource with no AS into it and you're
1221 using Full model, then ... you'll loose all interest into it. On the
1222 other hand, designing a binary tree of AS with, at the lower level,
1223 only one host, then you'll also loose all the good AS hierarchy can
1224 give you. Remind you should always be "reasonable" in your platform
1225 definition when choosing the hierarchy. A good choice if you try to
1226 describe a real life platform is to follow the AS described in
1227 reality, since this kind of trade-off works well for real life
1230 \subsection pf_exit_as Exit AS: why and how
1231 Users that have looked at some of our platforms may have notice a
1232 non-intuitive schema ... Something like that :
1236 <AS id="AS_4" routing="Full">
1237 <AS id="exitAS_4" routing="Full">
1238 <router id="router_4"/>
1240 <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"/>
1241 <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"/>
1242 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1243 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1244 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1245 <ASroute src="cl_4_1"
1247 gw_src="c_4_1-cl_4_1_router"
1248 gw_dst="c_4_2-cl_4_2_router"
1250 <link_ctn id="4_1"/>
1251 <link_ctn id="bb_4"/>
1252 <link_ctn id="4_2"/>
1254 <ASroute src="cl_4_1"
1256 gw_src="c_4_1-cl_4_1_router"
1259 <link_ctn id="4_1"/>
1260 <link_ctn id="bb_4"/>
1262 <ASroute src="cl_4_2"
1264 gw_src="c_4_2-cl_4_2_router"
1267 <link_ctn id="4_2"/>
1268 <link_ctn id="bb_4"/>
1273 In the AS_4, you have an exitAS_4 defined, containing only one router,
1274 and routes defined to that AS from all other AS (as cluster is only a
1275 shortcut for an AS, see cluster description for details). If there was
1276 an upper AS, it would define routes to and from AS_4 with the gateway
1277 router_4. It's just because, as we did not allowed (for performances
1278 issues) to have routes from an AS to a single host/router, you have to
1279 enclose your gateway, when you have AS included in your AS, within an
1280 AS to define routes to it.
1282 \subsection pf_P2P_tags P2P or how to use coordinates
1283 SimGrid allows you to use some coordinated-based system, like vivaldi,
1284 to describe a platform. The main concept is that you have some peers
1285 that are located somewhere: this is the function of the
1286 <b>coordinates</b> of the \<peer\> or \<host\> tag. There's nothing
1287 complicated in using it, here is an example of it:
1290 <?xml version='1.0'?>
1291 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1292 <platform version="3">
1294 <config id="General">
1295 <prop id="network/coordinates" value="yes"></prop>
1297 <AS id="AS0" routing="Vivaldi">
1298 <host id="100030591" coordinates="25.5 9.4 1.4" power="1500000000.0" />
1299 <host id="100036570" coordinates="-12.7 -9.9 2.1" power="730000000.0" />
1301 <host id="100429957" coordinates="17.5 6.7 18.8" power="830000000.0" />
1306 Coordinates are then used to calculate latency between two hosts by
1307 calculating the euclidean distance between the two hosts coordinates.
1308 The results express the latency in ms.
1310 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between AS.
1311 That is for example what is commonly done when using peers (see Section \ref pf_peer).
1313 <?xml version='1.0'?>
1314 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1315 <platform version="3">
1317 <config id="General">
1318 <prop id="network/coordinates" value="yes"></prop>
1320 <AS id="AS0" routing="Vivaldi">
1321 <peer id="peer-0" coordinates="173.0 96.8 0.1" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1322 <peer id="peer-1" coordinates="247.0 57.3 0.6" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1323 <peer id="peer-2" coordinates="243.4 58.8 1.4" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1327 In such a case though, we connect the AS created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1328 This means that to route between AS1 and AS2, it will use the coordinates of router_AS1 and router_AS2.
1329 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1330 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.
1333 \subsection pf_wisely Choosing wisely the routing model to use
1336 Choosing wisely the routing model to use can significantly fasten your
1337 simulation/save your time when writing the platform/save tremendous
1338 disk space. Here is the list of available model and their
1339 characteristics (lookup : time to resolve a route):
1341 \li <b>Full</b>: Full routing data (fast, large memory requirements,
1343 \li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1344 lookup, lesser memory requirements, shortest path routing only).
1345 Calculates all routes at once at the beginning.
1346 \li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1347 lookup, small memory requirements, shortest path routing only).
1348 Calculates a route when necessary.
1349 \li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1350 fast lookup, small memory requirements, shortest path routing
1351 only). Same as Dijkstra, except it handles a cache for latest used
1353 \li <b>none</b>: No routing (usable with Constant network only).
1354 Defines that there is no routes, so if you try to determine a
1355 route without constant network within this AS, SimGrid will raise
1357 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1358 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1361 \subsection pf_switch Hey, I want to describe a switch but there is no switch tag !
1363 Actually we did not include switch tag, ok. But when you're trying to
1364 simulate a switch, the only major impact it has when you're using
1365 fluid model (and SimGrid uses fluid model unless you activate GTNetS,
1366 ns-3, or constant network mode) is the impact of the upper limit of
1367 the switch motherboard speed that will eventually be reached if you're
1368 using intensively your switch. So, the switch impact is similar to a
1369 link one. That's why we are used to describe a switch using a link tag
1370 (as a link is not an edge by a hyperedge, you can connect more than 2
1373 \subsection pf_platform_multipath How to express multipath routing in platform files?
1375 It is unfortunately impossible to express the fact that there is more
1376 than one routing path between two given hosts. Let's consider the
1377 following platform file:
1380 <route src="A" dst="B">
1383 <route src="B" dst="C">
1386 <route src="A" dst="C">
1391 Although it is perfectly valid, it does not mean that data traveling
1392 from A to C can either go directly (using link 3) or through B (using
1393 links 1 and 2). It simply means that the routing on the graph is not
1394 trivial, and that data do not following the shortest path in number of
1395 hops on this graph. Another way to say it is that there is no implicit
1396 in these routing descriptions. The system will only use the routes you
1397 declare (such as <route src="A" dst="C"><link_ctn
1398 id="3"/></route>), without trying to build new routes by aggregating
1401 You are also free to declare platform where the routing is not
1402 symmetric. For example, add the following to the previous file:
1405 <route src="C" dst="A">
1411 This makes sure that data from C to A go through B where data from A
1412 to C go directly. Don't worry about realism of such settings since
1413 we've seen ways more weird situation in real settings (in fact, that's
1414 the realism of very regular platforms which is questionable, but
1415 that's another story).
1417 \section pf_flexml_bypassing Bypassing the XML parser with your own C functions
1418 <b>NOTE THAT THIS DOCUMENTATION, WHILE STILL WORKING, IS STRONGLY DEPRECATED</b>
1420 So you want to bypass the XML files parser, uh? Maybe doing some parameter
1421 sweep experiments on your simulations or so? This is possible, and
1422 it's not even really difficult (well. Such a brutal idea could be
1423 harder to implement). Here is how it goes.
1425 For this, you have to first remember that the XML parsing in SimGrid is done
1426 using a tool called FleXML. Given a DTD, this gives a flex-based parser. If
1427 you want to bypass the parser, you need to provide some code mimicking what
1428 it does and replacing it in its interactions with the SURF code. So, let's
1429 have a look at these interactions.
1431 FleXML parser are close to classical SAX parsers. It means that a
1432 well-formed SimGrid platform XML file might result in the following
1435 - start "platform_description" with attribute version="2"
1436 - start "host" with attributes id="host1" power="1.0"
1438 - start "host" with attributes id="host2" power="2.0"
1440 - start "link" with ...
1442 - start "route" with ...
1443 - start "link_ctn" with ...
1446 - end "platform_description"
1448 The communication from the parser to the SURF code uses two means:
1449 Attributes get copied into some global variables, and a surf-provided
1450 function gets called by the parser for each event. For example, the event
1451 - start "host" with attributes id="host1" power="1.0"
1453 let the parser do something roughly equivalent to:
1455 strcpy(A_host_id,"host1");
1460 In SURF, we attach callbacks to the different events by initializing the
1461 pointer functions to some the right surf functions. Since there can be
1462 more than one callback attached to the same event (if more than one
1463 model is in use, for example), they are stored in a dynar. Example in
1464 workstation_ptask_L07.c:
1466 /* Adding callback functions */
1467 surf_parse_reset_parser();
1468 surfxml_add_callback(STag_surfxml_host_cb_list, &parse_cpu_init);
1469 surfxml_add_callback(STag_surfxml_prop_cb_list, &parse_properties);
1470 surfxml_add_callback(STag_surfxml_link_cb_list, &parse_link_init);
1471 surfxml_add_callback(STag_surfxml_route_cb_list, &parse_route_set_endpoints);
1472 surfxml_add_callback(ETag_surfxml_link_c_ctn_cb_list, &parse_route_elem);
1473 surfxml_add_callback(ETag_surfxml_route_cb_list, &parse_route_set_route);
1475 /* Parse the file */
1476 surf_parse_open(file);
1477 xbt_assert(!surf_parse(), "Parse error in %s", file);
1481 So, to bypass the FleXML parser, you need to write your own version of the
1482 surf_parse function, which should do the following:
1483 - Fill the A_<tag>_<attribute> variables with the wanted values
1484 - Call the corresponding STag_<tag>_fun function to simulate tag start
1485 - Call the corresponding ETag_<tag>_fun function to simulate tag end
1486 - (do the same for the next set of values, and loop)
1488 Then, tell SimGrid that you want to use your own "parser" instead of the stock one:
1490 surf_parse = surf_parse_bypass_environment;
1491 MSG_create_environment(NULL);
1492 surf_parse = surf_parse_bypass_application;
1493 MSG_launch_application(NULL);
1496 A set of macros are provided at the end of
1497 include/surf/surfxml_parse.h to ease the writing of the bypass
1498 functions. An example of this trick is distributed in the file
1499 examples/msg/masterslave/masterslave_bypass.c