1 /*! \page platform Platform Description
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 it, either by using the Lua console (
13 \ref MSG_Lua_funct) or, if you're using MSG, some of MSG's platform and
14 deployment functions (\ref msg_simulation). If you want to use this,
15 check the particular documentation. (You can also check the section
16 \ref pf_flexml_bypassing, however, this documentation is deprecated;
17 there is a new, but undocumented, way to do it properly).
18 \li You can use two XML files: one contains the platform description while
19 the other contains the deployment instructions.
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.dtd">http://simgrid.gforge.inria.fr/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 3.
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
83 There is also the ``<route>`` tag; this tag takes two attributes, ``src`` (source)
84 and ``dst`` (destination). Both source and destination must be valid identifiers
85 for routers (these will be introduced later). Contained by the ``<route>`` are
86 network links; these links must be used in order to communicate from the source
87 to the destination specified in the tag. Hence, a route merely describes
88 how to reach a router from another router.
91 More information and (code-)examples can be found in the Section \ref pf_rm.
94 An AS can also contain one or more AS; this allows you to
95 define the hierarchy of your platform.
97 Within each AS, the following types of resources exist:
98 \li <b>host</b>: a hostmachine; contains processors/cores etc.
99 \li <b>router</b>: a router or a gateway.
100 \li <b>link</b>: a link that defines a connection between two (or
101 more) resources. Every link has a bandwidth and a latency.
102 \li <b>cluster</b>: like a real cluster, contains many hosts
103 interconnected by some dedicated network.
105 Between these elements, a routing has to be defined. The AS is
106 supposed to be Autonomous, hence this has to be done at the AS level. The AS
107 handles two different types of entities (<b>host/router</b> and
108 <b>AS</b>); you are responsible to define routes between those elements,
109 otherwise entities will be unconnected and therefore unreachable from other
110 entities. Although several algorithms for routing are built into SimGrid (see
111 \ref pf_rm), you might encounter a case where you want to define
112 routes manually (for instance, due to specific requirements of your
115 There are three tags to use:
116 \li <b>ASroute</b>: to define routes between two <b>AS</b>
117 \li <b>route</b>: to define routes between two <b>host/router</b>
118 \li <b>bypassRoute</b>: to define routes between two <b>AS</b> that
119 will bypass default routing (as specified by the ``routing`` attribute
120 supplied to ``<AS>``, see above).
122 Here is an illustration of these concepts:
124 ![An illustration of an AS hierarchy. Here, AS1 contains 5 other AS' who in turn may contain other AS' as well.](AS_hierarchy.png)
125 Circles represent processing units and squares represent network routers. Bold
126 lines represent communication links. AS2 models the core of a national
127 network interconnecting a small flat cluster (AS4) and a larger
128 hierarchical cluster (AS5), a subset of a LAN (AS6), and a set of peers
129 scattered around the world (AS7).
131 \section pf_pftags Resource description
133 \subsection pf_As Platform: The \<AS\> tag
135 The concept of an AS was already outlined above (Section \ref pf_basics);
136 recall that the AS is so important because it groups other resources (such
137 as routers/hosts) together (in fact, these resources must be contained by
140 Available attributes :
142 Attribute name | Mandatory | Values | Description
143 --------------- | --------- | ------ | -----------
144 id | yes | String | The identifier of an AS; facilitates referring to this AS. ID must be unique.
145 routing | yes | Full\| Floyd\| Dijkstra\| DijkstraCache\| none\| Vivaldi\| Cluster | See Section \ref pf_rm for details.
150 <AS id="AS0" routing="Full">
151 <host id="host1" power="1000000000"/>
152 <host id="host2" power="1000000000"/>
153 <link id="link1" bandwidth="125000000" latency="0.000100"/>
154 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
158 In this example, AS0 contains two hosts (host1 and host2). The route
159 between the hosts goes through link1.
161 \subsection pf_Cr Computing resources: hosts, clusters and peers.
163 \subsubsection pf_host The tag <host/>
165 A <b>host</b> represents a computer/node card. Every host is able to execute
166 code and it can send and receive data to/from other hosts. Most importantly,
167 a host can contain more than 1 core.
171 Attribute name | Mandatory | Values | Description
172 --------------- | --------- | ------ | -----------
173 id | yes | String | The identifier of the host. facilitates referring to this AS.
174 power | yes | double (must be > 0.0) | Computational power of every core of this host in FLOPS. Must be larger than 0.0.
175 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 #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.
176 availability | no | int | <b>Specify if the percentage of power available.</b> (What? TODO)
177 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.
178 state | no | ON\|OFF<br/> (Default: ON) | Is this host running or not?
179 state_file | no | string | Same mechanism as availability_file.<br /> <b>Note:</b> The filename must be specified with your system's format.
180 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.
182 ### Possible children: ###
184 Tag name | Description | Documentation
185 ------------ | ----------- | -------------
186 <mount/> | Defines mounting points between some storage resource and the host. | \ref pf_sto_mo
187 <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
192 <host id="host1" power="1000000000"/>
193 <host id="host2" power="1000000000">
194 <prop id="color" value="blue"/>
195 <prop id="rendershape" value="square"/>
200 \anchor pf_host_dynamism
201 ### Expressing dynamism ###
203 SimGrid provides mechanisms to change a hosts' availability over
204 time, using the ``availability_file`` attribute to the ``<host>`` tag
205 and a separate text file whose syntax is exemplified below.
207 #### Adding a trace file ####
210 <platform version="1">
211 <host id="bob" power="500000000" availability_file="bob.trace" />
215 #### Example of "bob.trace" file ####
224 Let us begin to explain this example by looking at line 2. (Line 1 will become clear soon).
225 The first column describes points in time, in this case, time 0. The second column
226 describes the relative amount of power this host is able to deliver (relative
227 to the maximum performance specified in the ``<host>`` tag). (Clearly, the
228 second column needs to contain values that are not smaller than 0 and not larger than 1).
229 In this example, our host will deliver 500 Mflop/s at time 0, as 500 Mflop/s is the
230 maximum performance of this host. At time 11.0, it will
231 deliver half of its maximum performance, i.e., 250 Mflop/s until time 20.0 when it will
232 will start delivering 80\% of its power. In this example, this amounts to 400 Mflop/s.
234 Since the periodicity in line 1 was set to be 1.0, i.e., 1 timestep, this host will
235 continue to provide 500 Mflop/s from time 21. From time 32 it will provide 250 MFlop/s and so on.
237 ### Changing initial state ###
239 It is also possible to specify whether the host is up or down by setting the
240 ``state`` attribute to either <b>ON</b> (default value) or <b>OFF</b>.
242 #### Example: Expliciting the default value "ON" ####
245 <platform version="1">
246 <host id="bob" power="500000000" state="ON" />
250 If you want this host to be unavailable, simply substitute ON with OFF.
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="1">
263 <host id="bob" power="500000000" state_file="bob.fail" />
267 #### Example of "bob.fail" file ####
275 A negative 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. 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 ``<cluster />`` represents a machine-cluster. Most of the time it is used
286 when you want to define many machines quickly. Technically,
287 ``cluster`` is a 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
330 Attribute name | Mandatory | Values | Description
331 --------------- | --------- | ------ | -----------
332 id | yes | string | The identifier of the cluster. Facilitates referring to this cluster.
333 prefix | yes | string | Each node of the cluster has to have a name. This name will be prefixed with this prefix.
334 suffix | yes | string | Each node of the cluster will be suffixed with this suffix
335 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.
336 power | yes | int | Same as the ``power`` attribute of the ``<host>`` tag.
337 core | no | int (default: 1) | Same as the ``core`` attribute of the ``<host>`` tag.
338 bw | yes | int | Bandwidth for the links between nodes and backbone (if any). <b>See <b>link</b> section for syntax/details.</b>
339 lat | yes | int | Latency for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
340 sharing_policy | no | string | Sharing policy for the links between nodes and backbone (if any). See <b>link</b> section for syntax/details.
341 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>).
342 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>).
343 bb_sharing_policy | no | string | Sharing policy for the backbone (if any). See <b>link</b> section for syntax/details.
344 availability_file | no | string | Allows you to use a file as input for availability. Similar to <b>hosts</b> attribute.
345 state_file | no | string | Allows you to use a file as input for states. Similar to <b>hosts</b> attribute.
346 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 <b>FATPIPE</b>.
347 loopback_lat | no | int | Latency for loopback (if any). See <b>link</b> section for syntax/details. See loopback_bw for more info.
348 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.
349 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 AsClusterFatTree "AsClusterFatTree documentation".
353 the router name is defined as the resulting String in the following
357 router_name = prefix + clusterId + _router + suffix;
361 #### Cluster example ####
363 Consider the following two (and independent) uses of the ``cluster`` tag:
366 <cluster id="my_cluster_1" prefix="" suffix="" radical="0-262144"
367 power="1e9" bw="125e6" lat="5E-5"/>
369 <cluster id="my_cluster_2" prefix="c-" suffix=".me" radical="0-99"
370 power="1e9" bw="125e6" lat="5E-5"
371 bb_bw="2.25e9" bb_lat="5E-4"/>
374 The second example creates one router and 100 machines with the following names:
376 c-my_cluster_2_router.me
384 \subsubsection pf_peer <peer/>
386 This tag represents a peer, as in Peer-to-Peer (P2P) networks. However, internally,
387 SimGrid transforms a peer into an AS (similar to Cluster). Hence, this tag
388 is virtually only a shortcut that comes with some pre-defined resources
389 and values. These are:
391 \li A tiny AS whose routing type is cluster is created
393 \li Two links: One for download and one for upload. This is
394 convenient to use and simulate stuff under the last mile model (e.g., ADSL peers).
395 \li It connects the two links to the host
396 \li It creates a router (a gateway) that serves as an entry point for this peer zone.
397 This router has coordinates.
401 Attribute name | Mandatory | Values | Description
402 --------------- | --------- | ------ | -----------
403 id | yes | string | The identifier of the peer. Facilitates referring to this peer.
404 power | yes | int | See the description of the ``host`` tag for this attribute
405 bw_in | yes | int | Bandwidth downstream
406 bw_out | yes | int | Bandwidth upstream
407 lat | yes | double | Latency for both up- and downstream, in seconds.
408 coordinates | no | string | Coordinates of the gateway for this peer. Example value: 12.8 14.4 6.4
409 sharing_policy | no | SHARED\|FULLDUPLEX (default: FULLDUPLEX) | Sharing policy for links. See <b>link</b> description for details.
410 availability_file| no | string | Availability file for the peer. Same as host availability file. See <b>host</b> description for details.
411 state_file | no | string | State file for the peer. Same as host state file. See <b>host</b> description for details.
413 Internally, SimGrid transforms any ``<peer/>`` construct such as
416 coordinates="12.8 14.4 6.4"
422 into an ``<AS>`` (see Sections \ref pf_basics and \ref pf_As). In fact, this example of the ``<peer/>`` tag
423 is completely equivalent to the following declaration:
426 <AS id="as_FOO" routing="Cluster">
427 <host id="peer_FOO" power="1.5Gf"/>
428 <link id="link_FOO_UP" bandwidth="2.25GBps" latency="500us"/>
429 <link id="link_FOO_DOWN" bandwidth="2.25GBps" latency="500us"/>
430 <router id="router_FOO" coordinates="25.5 9.4 1.4"/>
431 <host_link id="peer_FOO" up="link_FOO_UP" down="link_FOO_DOWN"/>
436 \subsection pf_ne Network equipments: links and routers
438 There are two tags available to represent network entities:
439 1. ``<link>``: Represents a entity that has a limited bandwidth, a
440 latency, and that can be shared according to TCP way to share this
443 The concept of links in SimGrid may not be intuitive, as links are not limited
444 to connecting (exactly) two entities; in fact, you can have more than two equipments
445 connected to it. (In graph theoretical terms: A link in SimGrid is not an edge,
449 2. ``<router/>``: Represents an entity that a message can be routed
450 to, but that is unable to execute any code. In SimGrid, routers have also
451 no impact on the performance: Routers do not limit any bandwidth nor
452 do they increase latency. As a matter of fact, routers are (almost) ignored
453 by the simulator when the simulation has begun.
456 If you want to represent an entity like a switch, you must use ``<link>`` (see section). Routers are used
457 to run some routing algorithm and determine routes (see Section \ref pf_routing for details).
460 \subsubsection pf_router <router/>
462 %As said before, <b>router</b> is used only to give some information
463 for routing algorithms. So, it does not have any attributes except :
467 Attribute name | Mandatory | Values | Description
468 --------------- | --------- | ------ | -----------
469 id | yes | string | The identifier of the router to be used when referring to it.
470 coordinates | yes | 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.
475 <router id="gw_dc1_horizdist"/>
478 \subsubsection pf_link <link>
480 Network links can represent one-hop network connections. They are
481 characterized by their id and their bandwidth; links can (but may not) be subject
486 Attribute name | Mandatory | Values | Description
487 --------------- | --------- | ------ | -----------
488 id | yes | string | The identifier of the link to be used when referring to it.
489 bandwidth | yes | int | Maximum bandwidth for this link, given in bytes/s
490 latency | no | double (default: 0.0) | Latency for this link.
491 sharing_policy | no | SHARED\|FATPIPE\|FULLDUPLEX (default: SHARED) | Sharing policy for the link.
492 state | no | ON\|OFF (default: ON) | Allows you to to turn this link on or off (working / not working)
493 bandwidth_file | no | string | Allows you to use a file as input for bandwidth.
494 latency_file | no | string | Allows you to use a file as input for latency.
495 state_file | no | string | Allows you to use a file as input for states.
498 #### Possible shortcuts for ``latency`` ####
500 When using the latency attribute, you can specify the latency by using the scientific
501 notation or by using common abbreviations. For instance, the following three tags
505 <link id="LINK1" bandwidth="125000000" latency="5E-6"/>
506 <link id="LINK1" bandwidth="125000000" latency="5us"/>
507 <link id="LINK1" bandwidth="125000000" latency="0.000005"/>
510 Here, the second tag uses "us", meaning "microseconds". Other shortcuts are:
512 Name | Abbreviation | Time (in seconds)
513 ---- | ------------ | -----------------
514 Week | w | 7 * 24 * 60 * 60
515 Day | d | 24 * 60 * 60
519 Millisecond | ms | 0.001 = 10^(-3)
520 Microsecond | us | 0.000001 = 10^(-6)
521 Nanosecond | ns | 0.000000001 = 10^(-9)
522 Picosecond | ps | 0.000000000001 = 10^(-12)
524 #### Sharing policy ####
526 By default a network link is SHARED, i.e., if two or more data flows go
527 through a link, the bandwidth is shared fairly among the data flows. This
528 is similar to the sharing policy TCP uses.
530 On the other hand, if a link is defined as a FATPIPE,
531 each flow going through this link will be allocated the complete bandwidth,
532 i.e., no sharing occurs and the bandwidth is only limiting single flows:
533 The complete bandwidth provided by this link in this mode
534 is ``#flows*bandwidth``.
536 The last mode available is FULLDUPLEX. This means that SimGrid will
537 automatically generate two links (one carrying the suffix _UP and the other the
538 suffix _DOWN) for each ``<link>`` tag. This models situations when the direction
539 of traffic is important.
542 Transfers from one side to the other will interact similarly as
543 TCP when ACK returning packets circulate on the other direction. More
544 discussion about it is available in the description of link_ctn description.
547 In other words: The SHARED policy defines a physical limit for the bandwidth.
548 The FATPIPE mode defines a limit for each application,
549 with no upper total limit.
552 Tip: By using the FATPIPE mode, you can model big backbones that
553 won't affect performance (except latency).
559 <link id="SWITCH" bandwidth="125000000" latency="5E-5" sharing_policy="FATPIPE" />
562 #### Expressing dynamism and failures ####
564 Similar to hosts, it is possible to declare links whose state, bandwidth
565 or latency changes over time (see Section \ref pf_hosts_dynamism for details).
567 In the case of network links, the ``bandwidth`` and ``latency`` attributes are
568 replaced by the ``bandwidth_file`` and ``latency_file`` attributes.
569 The following XML snippet demonstrates how to use this feature in the platform
570 file. The structure of the files "link1.bw" and "link1.lat" is shown below.
573 <link id="LINK1" state_file="link1.fail" bandwidth="80000000" latency=".0001" bandwidth_file="link1.bw" latency_file="link1.lat" />
577 Even if the syntax is the same, the semantic of bandwidth and latency
578 trace files differs from that of host availability files. For bandwidth and
579 latency, the corresponding files do not
580 express availability as a fraction of the available capacity but directly in
581 bytes per seconds for the bandwidth and in seconds for the latency. This is
582 because most tools allowing to capture traces on real platforms (such as NWS)
583 express their results this way.
586 ##### Example of "link1.bw" file #####
594 In this example, the bandwidth changes repeatedly, with all changes
595 being repeated every 12 seconds.
597 At the beginning of the the simulation, the link's bandwidth is 80,000,000
598 B/s (i.e., 80 Mb/s); this value was defined in the XML snippet above.
599 After four seconds, it drops to 40 Mb/s (line 2), and climbs
600 back to 60 Mb/s after another 4 seconds (line 3). The value does not change any
601 more until the end of the period, that is, after 12 seconds have been simulated).
602 At this point, periodicity kicks in and this behavior is repeated: Seconds
603 12-16 will experience 80 Mb/s, 16-20 40 Mb/s etc.).
605 ##### Example of "link1.lat" file #####
614 In this example, the latency varies with a period of 5 seconds.
615 In the xml snippet above, the latency is initialized to be 0.0001s (100µs). This
616 value will be kept during the first second, since the latency_file contains
617 changes to this value at second one, two and three.
618 At second one, the value will be 0.001, i.e., 1ms. One second later it will
619 be adjusted to 0.01 (or 10ms) and one second later it will be set again to 1ms. The
620 value will not change until second 5, when the periodicity defined in line 1
621 kicks in. It then loops back, starting at 100µs (the initial value) for one second.
624 #### The ``<prop/>`` tag ####
626 Similar to ``<host>``, the link may also contain the ``<prop/>`` tag; see the host
627 documentation (Section \ref pf_host) for an example.
632 \subsection pf_storage Storage
635 This is a prototype version that should evolve quickly, this
636 is just some doc valuable only at the time of writing this doc</b>
637 This section describes the storage management under SimGrid ; nowadays
638 it's only usable with MSG. It relies basically on linux-like concepts.
639 You also may want to have a look to its corresponding section in \ref
640 msg_file_management ; functions access are organized as a POSIX-like
644 \subsubsection pf_sto_conc Storage Main concepts
645 Basically there is 3 different entities to know :
646 \li the <b>storage_type</b>: here you define some kind of storage that
647 you will instantiate many type on your platform. Think of it like
648 a definition of throughput of a specific disk.
649 \li the <b>storage</b>: instance of a <b>storage_type</b>. Defines a
650 new storage of <b>storage_type</b>
651 \li the <b>mount</b>: says that the storage is located into this
654 the content of a storage has to be defined in a content file that
655 contains the content. The path to this file has to be passed within
656 the <b>content</b> attribute . Here is a way to generate it:
659 find /path/you/want -type f -exec ls -l {} \; 2>/dev/null > ./content.txt
662 \subsubsection pf_sto_sttp storage_type
665 <b>storage_type</b> attributes :
666 \li <b>id (mandatory)</b>: the identifier of the storage_type to be
667 used when referring to it.
668 \li <b>model (mandatory)</b>: Unused for now by the simulator (but
670 \li <b>content</b>: default value 0. The file containing the disk
671 content. (may be moved soon or later to <b>storage</b> tag.
673 The tag must contains some predefined model prop, as may do some other
675 <b>storage_type</b> mandatory <b>model_prop</b> :
676 \li <b>Bwrite</b>: value in B/s. Write throughput
677 \li <b>Bread</b>: value in B/s. Read throughput
678 \li <b>Bconnexion</b>: value in B/s. Connection throughput (i.e. the
679 throughput of the storage connector).
681 A storage_type can also contain the <b>prop</b> tag. The prop tag allows you
682 to define additional information on this storage_type following the
683 attribute/value schema. You may want to use it to give information to
684 the tool you use for rendering your simulation, for example.
687 <storage_type id="single_HDD" model="linear_no_lat" size="4000" content_type="txt_unix">
688 <model_prop id="Bwrite" value="30MBps" />
689 <model_prop id="Bread" value="100MBps" />
690 <model_prop id="Bconnection" value="150MBps" />
691 <b><prop id="Brand" value="Western Digital" /></b>
695 \subsubsection pf_sto_st storage
697 <b>storage_type</b> attributes :
698 \li <b>id (mandatory)</b>: the identifier of the storage to be used
699 when referring to it.
700 \li <b>typeId (mandatory)</b>: the identifier of the storage_type that
701 this storage belongs to.
702 \li <b>attach (mandatory)</b>: the host (name) to which the storage is
705 \subsubsection pf_sto_mo mount
707 <b>mount</b> attributes :
708 \li <b>id (mandatory)</b>: the id of the <b>storage</b> that must be
709 mounted on that computer.
710 \li <b>name (mandatory)</b>: the name that will be the logical
711 reference to this disk (the mount point).
713 \subsubsection pf_sto_mst mstorage
714 <b>Note : unused for now</b>
715 <b>mstorage</b> attributes :
716 \li <b>typeId (mandatory)</b>: the id of the <b>storage</b> that must
717 be mounted on that computer.
718 \li <b>name (mandatory)</b>: the name that will be the logical
719 reference to this disk (the mount point).
721 \section pf_routing Routing
723 To achieve high performance, the routing tables used within SimGrid are
724 static. This means that routing between two nodes is calculated once
725 and will not change during execution. The SimGrid team chose to use this
726 approach as it is rare to have a real deficiency of a resource;
727 most of the time, a communication fails because the links experience too much
728 congestion and hence, your connection stops before the timeout or
729 because the computer designated to be the destination of that message
732 We also chose to use shortest paths algorithms in order to emulate
733 routing. Doing so is consistent with the reality: RIP, OSPF, BGP are
734 all calculating shortest paths. They have some convergence time, but
735 at the end, so when the platform is stable (and this should be the
736 moment you want to simulate something using SimGrid) your packets will
737 follow the shortest paths.
739 \subsection pf_rm Routing models
741 Within each AS, you have to define a routing model to use. You have
742 basically 3 main kind of routing models :
744 \li Shortest-path based models: you let SimGrid calculates shortest
745 paths and manage it. Behaves more or less as most real life
747 \li Manually-entered route models: you'll have to define all routes
748 manually by yourself into the platform description file.
749 Consistent with some manually managed real life routing.
750 \li Simple/fast models: those models offers fast, low memory routing
751 algorithms. You should consider to use it if you can make some
752 assumptions about your AS. Routing in this case is more or less
755 \subsubsection pf_raf The router affair
757 Expressing routers becomes mandatory when using shortest-path based
758 models or when using ns-3 or the bindings to the GTNetS packet-level
759 simulator instead of the native analytical network model implemented
762 For graph-based shortest path algorithms, routers are mandatory,
763 because both algorithms need a graph, and so we need to have source
764 and destination for each edge.
766 Routers are naturally an important concept in GTNetS or ns-3 since the
767 way they run the packet routing algorithms is actually simulated.
768 Instead, the SimGrid’s analytical models aggregate the routing time
769 with the transfer time. Rebuilding a graph representation only from
770 the route information turns to be a very difficult task, because of
771 the missing information about how routes intersect. That is why we
772 introduced a \<router\> tag, which is simply used to express these
773 intersection points. The only attribute accepted by this tag an id. It
774 is important to understand that the \<router\> tag is only used to
775 provide topological information.
777 To express those topological information, some <b>route</b> have to be
778 defined saying which link is between which routers. Description or the
779 route syntax is given below, as well as example for the different
782 \subsubsection pf_rm_sh Shortest-path based models
784 Here is the complete list of such models, that computes routes using
785 classic shortest-paths algorithms. How to choose the best suited
786 algorithm is discussed later in the section devoted to it.
788 \li <b>Floyd</b>: Floyd routing data. Pre-calculates all routes once.
789 \li <b>Dijkstra</b>: Dijkstra routing data ,calculating routes when
791 \li <b>DijkstraCache</b>: Dijkstra routing data. Handle some cache for
792 already calculated routes.
794 All those shortest-path models are instanciated the same way. Here are
799 <AS id="AS0" routing="Floyd">
801 <cluster id="my_cluster_1" prefix="c-" suffix=""
802 radical="0-1" power="1000000000" bw="125000000" lat="5E-5"
803 router_id="router1"/>
805 <AS id="AS1" routing="none">
806 <host id="host1" power="1000000000"/>
809 <link id="link1" bandwidth="100000" latency="0.01"/>
811 <ASroute src="my_cluster_1" dst="AS1"
814 <link_ctn id="link1"/>
820 ASroute given at the end gives a topological information: link1 is
821 between router1 and host1.
825 <AS id="AS_2" routing="Dijsktra">
826 <host id="AS_2_host1" power="1000000000"/>
827 <host id="AS_2_host2" power="1000000000"/>
828 <host id="AS_2_host3" power="1000000000"/>
829 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
830 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
831 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
832 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
833 <router id="central_router"/>
834 <router id="AS_2_gateway"/>
835 <!-- routes providing topological information -->
836 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
837 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
838 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
839 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
843 DijsktraCache example :
845 <AS id="AS_2" routing="DijsktraCache">
846 <host id="AS_2_host1" power="1000000000"/>
848 (platform unchanged compared to upper example)
851 \subsubsection pf_rm_me Manually-entered route models
853 \li <b>Full</b>: You have to enter all necessary routes manually
857 <AS id="AS0" routing="Full">
858 <host id="host1" power="1000000000"/>
859 <host id="host2" power="1000000000"/>
860 <link id="link1" bandwidth="125000000" latency="0.000100"/>
861 <route src="host1" dst="host2"><link_ctn id="link1"/></route>
865 \subsubsection pf_rm_sf Simple/fast models
867 \li <b>none</b>: No routing (Unless you know what you are doing, avoid
868 using this mode in combination with a non Constant network model).
871 <AS id="exitAS" routing="none">
872 <router id="exit_gateway"/>
875 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use
876 coordinates. See the corresponding section P2P below for details.
877 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
878 not be used, except internally.
880 \subsection ps_dec Defining routes
882 The principle of route definition is the same for the 4 available tags
883 for doing it. Those for tags are:
885 \li <b>route</b>: to define route between host/router
886 \li <b>ASroute</b>: to define route between AS
887 \li <b>bypassRoute</b>: to bypass normal routes as calculated by the
888 network model between host/router
889 \li <b>bypassASroute</b>: same as bypassRoute, but for AS
891 Basically all those tags will contain an (ordered) list of references
892 to link that compose the route you want to define.
894 Consider the example below:
897 <route src="Alice" dst="Bob">
898 <link_ctn id="link1"/>
899 <link_ctn id="link2"/>
900 <link_ctn id="link3"/>
904 The route here from host Alice to Bob will be first link1, then link2,
905 and finally link3. What about the reverse route ? <b>route</b> and
906 <b>ASroute</b> have an optional attribute <b>symmetrical</b>, that can
907 be either YES or NO. YES means that the reverse route is the same
908 route in the inverse order, and is set to YES by default. Note that
909 this is not the case for bypass*Route, as it is more probable that you
910 want to bypass only one default route.
912 For an ASroute, things are just slightly more complicated, as you have
913 to give the id of the gateway which is inside the AS you're talking
914 about you want to access ... So it looks like this :
918 <ASroute src="AS1" dst="AS2"
919 gw_src="router1" gw_dst="router2">
920 <link_ctn id="link1"/>
924 gw == gateway, so when any message are trying to go from AS1 to AS2,
925 it means that it must pass through router1 to get out of the AS, then
926 pass through link1, and get into AS2 by being received by router2.
927 router1 must belong to AS1 and router2 must belong to AS2.
929 \subsubsection pf_linkctn link_ctn
931 a <b>link_ctn</b> is the tag that is used in order to reference a
932 <b>link</b> in a route. Its id is the link id it refers to.
934 <b>link_ctn</b> attributes :
935 \li <b>id (mandatory)</b>: Id of the link this tag refers to
936 \li <b>direction</b>: if the link referenced by <b>id</b> has been
937 declared as FULLDUPLEX, this is used to indicate in which
938 direction the route you're defining is going through this link.
939 Possible values "UP" or "DOWN".
941 \subsubsection pf_asro ASroute
943 ASroute tag purpose is to let people write manually their routes
944 between AS. It's useful when you're in Full model.
946 <b>ASroute</b> attributes :
947 \li <b>src (mandatory)</b>: the source AS id.
948 \li <b>dst (mandatory)</b>: the destination AS id.
949 \li <b>gw_src (mandatory)</b>: the gateway to be used within the AS.
950 Can be any <b>host</b> or \b router defined into the \b src AS or
951 into one of the AS it includes.
952 \li <b>gw_dst (mandatory)</b>: the gateway to be used within the AS.
953 Can be any <b>host</b> or \b router defined into the \b dst AS or
954 into one of the AS it includes.
955 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
956 will be the opposite of the one defined. Can be either YES or NO,
959 <b>Example of ASroute with Full</b>
961 <AS id="AS0" routing="Full">
962 <cluster id="my_cluster_1" prefix="c-" suffix=".me"
963 radical="0-149" power="1000000000" bw="125000000" lat="5E-5"
964 bb_bw="2250000000" bb_lat="5E-4"/>
966 <cluster id="my_cluster_2" prefix="c-" suffix=".me"
967 radical="150-299" power="1000000000" bw="125000000" lat="5E-5"
968 bb_bw="2250000000" bb_lat="5E-4"/>
970 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
972 <ASroute src="my_cluster_1" dst="my_cluster_2"
973 gw_src="c-my_cluster_1_router.me"
974 gw_dst="c-my_cluster_2_router.me">
975 <link_ctn id="backbone"/>
977 <ASroute src="my_cluster_2" dst="my_cluster_1"
978 gw_src="c-my_cluster_2_router.me"
979 gw_dst="c-my_cluster_1_router.me">
980 <link_ctn id="backbone"/>
985 \subsubsection pf_ro route
986 The principle is the same as ASroute : <b>route</b> contains list of
987 links that are in the path between src and dst, except that it is for
988 routes between a src that can be either <b>host</b> or \b router and a
989 dst that can be either <b>host</b> or \b router. Useful for Full
990 as well as for the shortest-paths based models, where you
991 have to give topological information.
994 <b>route</b> attributes :
995 \li <b>src (mandatory)</b>: the source id.
996 \li <b>dst (mandatory)</b>: the destination id.
997 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
998 will be the opposite of the one defined. Can be either YES or NO,
1001 <b>route example in Full</b>
1003 <route src="Tremblay" dst="Bourassa">
1004 <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"/>
1008 <b>route example in a shortest-path model</b>
1010 <route src="Tremblay" dst="Bourassa">
1014 Note that when using route to give topological information, you have
1015 to give routes with one link only in it, as SimGrid needs to know
1016 which host are at the end of the link.
1018 \subsubsection pf_byASro bypassASroute
1020 <b>Note : bypassASroute and bypassRoute are under rewriting to perform
1021 better ; so you may not use it yet</b> As said before, once you choose
1022 a model, it (if so) calculates routes for you. But maybe you want to
1023 define some of your routes, which will be specific. You may also want
1024 to bypass some routes defined in lower level AS at an upper stage :
1025 <b>bypassASroute</b> is the tag you're looking for. It allows to
1026 bypass routes defined between already defined between AS (if you want
1027 to bypass route for a specific host, you should just use byPassRoute).
1028 The principle is the same as ASroute : <b>bypassASroute</b> contains
1029 list of links that are in the path between src and dst.
1031 <b>bypassASroute</b> attributes :
1032 \li <b>src (mandatory)</b>: the source AS id.
1033 \li <b>dst (mandatory)</b>: the destination AS id.
1034 \li <b>gw_src (mandatory)</b>: the gateway to be used within the AS.
1035 Can be any <b>host</b> or \b router defined into the \b src AS or
1036 into one of the AS it includes.
1037 \li <b>gw_dst (mandatory)</b>: the gateway to be used within the AS.
1038 Can be any <b>host</b> or \b router defined into the \b dst AS or
1039 into one of the AS it includes.
1040 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
1041 will be the opposite of the one defined. Can be either YES or NO,
1044 <b>bypassASroute Example</b>
1046 <bypassASRoute src="my_cluster_1" dst="my_cluster_2"
1047 gw_src="my_cluster_1_router"
1048 gw_dst="my_cluster_2_router">
1049 <link_ctn id="link_tmp"/>
1053 \subsubsection pf_byro bypassRoute
1054 <b>Note : bypassASRoute and bypassRoute are under rewriting to perform
1055 better ; so you may not use it yet</b> As said before, once you choose
1056 a model, it (if so) calculates routes for you. But maybe you want to
1057 define some of your routes, which will be specific. You may also want
1058 to bypass some routes defined in lower level AS at an upper stage :
1059 <b>bypassRoute</b> is the tag you're looking for. It allows to bypass
1060 routes defined between <b>host/router</b>. The principle is the same
1061 as route : <b>bypassRoute</b> contains list of links references of
1062 links that are in the path between src and dst.
1064 <b>bypassRoute</b> attributes :
1065 \li <b>src (mandatory)</b>: the source AS id.
1066 \li <b>dst (mandatory)</b>: the destination AS id.
1067 \li <b>symmetrical</b>: if the route is symmetric, the reverse route
1068 will be the opposite of the one defined. Can be either YES or NO,
1071 <b>bypassRoute Example</b>
1073 <bypassRoute src="host_1" dst="host_2">
1074 <link_ctn id="link_tmp"/>
1079 \subsection pb_baroex Basic Routing Example
1081 Let's say you have an AS named AS_Big that contains two other AS, AS_1
1082 and AS_2. If you want to make a host (h1) from AS_1 with another one
1083 (h2) from AS_2 then you'll have to proceed as follows:
1084 \li First, you have to ensure that a route is defined from h1 to the
1085 AS_1's exit gateway and from h2 to AS_2's exit gateway.
1086 \li Then, you'll have to define a route between AS_1 to AS_2. As those
1087 AS are both resources belonging to AS_Big, then it has to be done
1088 at AS_big level. To define such a route, you have to give the
1089 source AS (AS_1), the destination AS (AS_2), and their respective
1090 gateway (as the route is effectively defined between those two
1091 entry/exit points). Elements of this route can only be elements
1092 belonging to AS_Big, so links and routers in this route should be
1093 defined inside AS_Big. If you choose some shortest-path model,
1094 this route will be computed automatically.
1096 As said before, there are mainly 2 tags for routing :
1097 \li <b>ASroute</b>: to define routes between two <b>AS</b>
1098 \li <b>route</b>: to define routes between two <b>host/router</b>
1100 As we are dealing with routes between AS, it means that those we'll
1101 have some definition at AS_Big level. Let consider AS_1 contains 1
1102 host, 1 link and one router and AS_2 3 hosts, 4 links and one router.
1103 There will be a central router, and a cross-like topology. At the end
1104 of the crosses arms, you'll find the 3 hosts and the router that will
1105 act as a gateway. We have to define routes inside those two AS. Let
1106 say that AS_1 contains full routes, and AS_2 contains some Floyd
1107 routing (as we don't want to bother with defining all routes). As
1108 we're using some shortest path algorithms to route into AS_2, we'll
1109 then have to define some <b>route</b> to gives some topological
1110 information to SimGrid. Here is a file doing it all :
1113 <AS id="AS_Big" routing="Dijsktra">
1114 <AS id="AS_1" routing="Full">
1115 <host id="AS_1_host1" power="1000000000"/>
1116 <link id="AS_1_link" bandwidth="1250000000" latency="5E-4"/>
1117 <router id="AS_1_gateway"/>
1118 <route src="AS_1_host1" dst="AS_1_gateway">
1119 <link_ctn id="AS_1_link"/>
1122 <AS id="AS_2" routing="Floyd">
1123 <host id="AS_2_host1" power="1000000000"/>
1124 <host id="AS_2_host2" power="1000000000"/>
1125 <host id="AS_2_host3" power="1000000000"/>
1126 <link id="AS_2_link1" bandwidth="1250000000" latency="5E-4"/>
1127 <link id="AS_2_link2" bandwidth="1250000000" latency="5E-4"/>
1128 <link id="AS_2_link3" bandwidth="1250000000" latency="5E-4"/>
1129 <link id="AS_2_link4" bandwidth="1250000000" latency="5E-4"/>
1130 <router id="central_router"/>
1131 <router id="AS_2_gateway"/>
1132 <!-- routes providing topological information -->
1133 <route src="central_router" dst="AS_2_host1"><link_ctn id="AS_2_link1"/></route>
1134 <route src="central_router" dst="AS_2_host2"><link_ctn id="AS_2_link2"/></route>
1135 <route src="central_router" dst="AS_2_host3"><link_ctn id="AS_2_link3"/></route>
1136 <route src="central_router" dst="AS_2_gateway"><link_ctn id="AS_2_link4"/></route>
1138 <link id="backbone" bandwidth="1250000000" latency="5E-4"/>
1140 <ASroute src="AS_1" dst="AS_2"
1141 gw_src="AS_1_gateway"
1142 gw_dst="AS_2_gateway">
1143 <link_ctn id="backbone"/>
1148 \section pf_other_tags Tags not (directly) describing the platform
1150 There are 3 tags, that you can use inside a \<platform\> tag that are
1151 not describing the platform:
1152 \li random: it allows you to define random generators you want to use
1153 for your simulation.
1154 \li config: it allows you to pass some configuration stuff like, for
1155 example, the network model and so on. It follows the
1156 \li include: simply allows you to include another file into the
1159 \subsection pf_conf config
1160 <b>config</b> attributes :
1161 \li <b>id (mandatory)</b>: the identifier of the config to be used
1162 when referring to it.
1165 <b>config</b> tag only purpose is to include <b>prop</b> tags. Valid
1166 id are basically the same as the list of possible parameters you can
1167 use by command line, except that "/" are used for namespace
1168 definition. See the \ref options config and options page for more
1172 <b>config example</b>
1174 <?xml version='1.0'?>
1175 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1176 <platform version="3">
1177 <config id="General">
1178 <prop id="maxmin/precision" value="0.000010"></prop>
1179 <prop id="cpu/optim" value="TI"></prop>
1180 <prop id="workstation/model" value="compound"></prop>
1181 <prop id="network/model" value="SMPI"></prop>
1182 <prop id="path" value="~/"></prop>
1183 <prop id="smpi/bw_factor" value="65472:0.940694;15424:0.697866;9376:0.58729"></prop>
1186 <AS id="AS0" routing="Full">
1191 \subsection pf_rand random
1192 Not yet in use, and possibly subject to huge modifications.
1194 \subsection pf_incl include
1195 <b>include</b> tag allows to import into a file platform parts located
1196 in another file. This is done with the intention to help people
1197 combine their different AS and provide new platforms. Those files
1198 should contains XML part that contains either
1199 <b>include,cluster,peer,AS,trace,trace_connect</b> tags.
1201 <b>include</b> attributes :
1202 \li <b>file (mandatory)</b>: filename of the file to include. Possible
1203 values: absolute or relative path, syntax similar to the one in
1206 <b>Note</b>: due to some obscure technical reasons, you have to open
1207 and close tag in order to let it work.
1208 <b>include Example</b>
1210 <?xml version='1.0'?>
1211 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1212 <platform version="3">
1213 <AS id="main" routing="Full">
1214 <include file="clusterA.xml"></include>
1215 <include file="clusterB.xml"></include>
1220 \subsection pf_tra trace and trace_connect
1221 Both tags are an alternate way to passe availability, state, and so on
1222 files to entity. Instead of referring to the file directly in the host,
1223 link, or cluster tag, you proceed by defining a trace with an id
1224 corresponding to a file, later a host/link/cluster, and finally using
1225 trace_connect you say that the file trace must be used by the entity.
1226 Get it ? Let's have a look at an example :
1229 <AS id="AS0" routing="Full">
1230 <host id="bob" power="1000000000"/>
1232 <trace id="myTrace" file="bob.trace" periodicity="1.0"/>
1233 <trace_connect trace="myTrace" element="bob" kind="POWER"/>
1236 All constraints you have is that <b>trace_connect</b> is after
1237 <b>trace</b> and <b>host</b> definitions.
1240 <b>trace</b> attributes :
1241 \li <b>id (mandatory)</b>: the identifier of the trace to be used when
1243 \li <b>file</b>: filename of the file to include. Possible values :
1244 absolute or relative path, syntax similar to the one in use on
1245 your system. If omitted, the system expects that you provide the
1246 trace values inside the trace tags (see below).
1247 \li <b>trace periodicity (mandatory)</b>: trace periodicity, same
1248 definition as in hosts (see upper for details).
1250 Here is an example of trace when no file name is provided:
1253 <trace id="myTrace" periodicity="1.0">
1260 <b>trace_connect</b> attributes :
1261 \li <b>kind</b>: the type of trace, possible values
1262 <b>HOST_AVAIL|POWER|LINK_AVAIL|BANDWIDTH|LATENCY,</b> default:
1264 \li <b>trace (mandatory)</b>: the identifier of the trace referenced.
1265 \li <b>element (mandatory)</b>: the identifier of the entity referenced.
1269 \section pf_hints Hints and tips, or how to write a platform efficiently
1271 Now you should know at least the syntax and be able to create a
1272 platform by your own. However, after having ourselves wrote some platforms, there
1273 are some best practices you should pay attention to in order to
1274 produce good platform and some choices you can make in order to have
1275 faster simulations. Here's some hints and tips, then.
1277 \subsection pf_as_h AS Hierarchy
1278 The AS design allows SimGrid to go fast, because computing route is
1279 done only for the set of resources defined in this AS. If you're using
1280 only a big AS containing all resource with no AS into it and you're
1281 using Full model, then ... you'll loose all interest into it. On the
1282 other hand, designing a binary tree of AS with, at the lower level,
1283 only one host, then you'll also loose all the good AS hierarchy can
1284 give you. Remind you should always be "reasonable" in your platform
1285 definition when choosing the hierarchy. A good choice if you try to
1286 describe a real life platform is to follow the AS described in
1287 reality, since this kind of trade-off works well for real life
1290 \subsection pf_exit_as Exit AS: why and how
1291 Users that have looked at some of our platforms may have notice a
1292 non-intuitive schema ... Something like that :
1296 <AS id="AS_4" routing="Full">
1297 <AS id="exitAS_4" routing="Full">
1298 <router id="router_4"/>
1300 <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"/>
1301 <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"/>
1302 <link id="4_1" bandwidth="2250000000" latency="5E-5"/>
1303 <link id="4_2" bandwidth="2250000000" latency="5E-5"/>
1304 <link id="bb_4" bandwidth="2250000000" latency="5E-4"/>
1305 <ASroute src="cl_4_1"
1307 gw_src="c_4_1-cl_4_1_router"
1308 gw_dst="c_4_2-cl_4_2_router"
1310 <link_ctn id="4_1"/>
1311 <link_ctn id="bb_4"/>
1312 <link_ctn id="4_2"/>
1314 <ASroute src="cl_4_1"
1316 gw_src="c_4_1-cl_4_1_router"
1319 <link_ctn id="4_1"/>
1320 <link_ctn id="bb_4"/>
1322 <ASroute src="cl_4_2"
1324 gw_src="c_4_2-cl_4_2_router"
1327 <link_ctn id="4_2"/>
1328 <link_ctn id="bb_4"/>
1333 In the AS_4, you have an exitAS_4 defined, containing only one router,
1334 and routes defined to that AS from all other AS (as cluster is only a
1335 shortcut for an AS, see cluster description for details). If there was
1336 an upper AS, it would define routes to and from AS_4 with the gateway
1337 router_4. It's just because, as we did not allowed (for performances
1338 issues) to have routes from an AS to a single host/router, you have to
1339 enclose your gateway, when you have AS included in your AS, within an
1340 AS to define routes to it.
1342 \subsection pf_P2P_tags P2P or how to use coordinates
1343 SimGrid allows you to use some coordinated-based system, like vivaldi,
1344 to describe a platform. The main concept is that you have some peers
1345 that are located somewhere: this is the function of the
1346 <b>coordinates</b> of the \<peer\> or \<host\> tag. There's nothing
1347 complicated in using it, here is an example of it:
1350 <?xml version='1.0'?>
1351 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1352 <platform version="3">
1354 <config id="General">
1355 <prop id="network/coordinates" value="yes"></prop>
1357 <AS id="AS0" routing="Vivaldi">
1358 <host id="100030591" coordinates="25.5 9.4 1.4" power="1500000000.0" />
1359 <host id="100036570" coordinates="-12.7 -9.9 2.1" power="730000000.0" />
1361 <host id="100429957" coordinates="17.5 6.7 18.8" power="830000000.0" />
1366 Coordinates are then used to calculate latency between two hosts by
1367 calculating the euclidean distance between the two hosts coordinates.
1368 The results express the latency in ms.
1370 Note that the previous example defines a routing directly between hosts but it could be also used to define a routing between AS.
1371 That is for example what is commonly done when using peers (see Section \ref pf_peer).
1373 <?xml version='1.0'?>
1374 <!DOCTYPE platform SYSTEM "http://simgrid.gforge.inria.fr/simgrid.dtd">
1375 <platform version="3">
1377 <config id="General">
1378 <prop id="network/coordinates" value="yes"></prop>
1380 <AS id="AS0" routing="Vivaldi">
1381 <peer id="peer-0" coordinates="173.0 96.8 0.1" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us"/>
1382 <peer id="peer-1" coordinates="247.0 57.3 0.6" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1383 <peer id="peer-2" coordinates="243.4 58.8 1.4" power="730Mf" bw_in="13.38MBps" bw_out="1.024MBps" lat="500us" />
1387 In such a case though, we connect the AS created by the <b>peer</b> tag with the Vivaldi routing mechanism.
1388 This means that to route between AS1 and AS2, it will use the coordinates of router_AS1 and router_AS2.
1389 This is currently a convention and we may offer to change this convention in the DTD later if needed.
1390 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.
1393 \subsection pf_wisely Choosing wisely the routing model to use
1396 Choosing wisely the routing model to use can significantly fasten your
1397 simulation/save your time when writing the platform/save tremendous
1398 disk space. Here is the list of available model and their
1399 characteristics (lookup : time to resolve a route):
1401 \li <b>Full</b>: Full routing data (fast, large memory requirements,
1403 \li <b>Floyd</b>: Floyd routing data (slow initialization, fast
1404 lookup, lesser memory requirements, shortest path routing only).
1405 Calculates all routes at once at the beginning.
1406 \li <b>Dijkstra</b>: Dijkstra routing data (fast initialization, slow
1407 lookup, small memory requirements, shortest path routing only).
1408 Calculates a route when necessary.
1409 \li <b>DijkstraCache</b>: Dijkstra routing data (fast initialization,
1410 fast lookup, small memory requirements, shortest path routing
1411 only). Same as Dijkstra, except it handles a cache for latest used
1413 \li <b>none</b>: No routing (usable with Constant network only).
1414 Defines that there is no routes, so if you try to determine a
1415 route without constant network within this AS, SimGrid will raise
1417 \li <b>Vivaldi</b>: Vivaldi routing, so when you want to use coordinates
1418 \li <b>Cluster</b>: Cluster routing, specific to cluster tag, should
1421 \subsection pf_switch Hey, I want to describe a switch but there is no switch tag !
1423 Actually we did not include switch tag, ok. But when you're trying to
1424 simulate a switch, the only major impact it has when you're using
1425 fluid model (and SimGrid uses fluid model unless you activate GTNetS,
1426 ns-3, or constant network mode) is the impact of the upper limit of
1427 the switch motherboard speed that will eventually be reached if you're
1428 using intensively your switch. So, the switch impact is similar to a
1429 link one. That's why we are used to describe a switch using a link tag
1430 (as a link is not an edge by a hyperedge, you can connect more than 2
1433 \subsection pf_platform_multipath How to express multipath routing in platform files?
1435 It is unfortunately impossible to express the fact that there is more
1436 than one routing path between two given hosts. Let's consider the
1437 following platform file:
1440 <route src="A" dst="B">
1443 <route src="B" dst="C">
1446 <route src="A" dst="C">
1451 Although it is perfectly valid, it does not mean that data traveling
1452 from A to C can either go directly (using link 3) or through B (using
1453 links 1 and 2). It simply means that the routing on the graph is not
1454 trivial, and that data do not following the shortest path in number of
1455 hops on this graph. Another way to say it is that there is no implicit
1456 in these routing descriptions. The system will only use the routes you
1457 declare (such as <route src="A" dst="C"><link_ctn
1458 id="3"/></route>), without trying to build new routes by aggregating
1461 You are also free to declare platform where the routing is not
1462 symmetric. For example, add the following to the previous file:
1465 <route src="C" dst="A">
1471 This makes sure that data from C to A go through B where data from A
1472 to C go directly. Don't worry about realism of such settings since
1473 we've seen ways more weird situation in real settings (in fact, that's
1474 the realism of very regular platforms which is questionable, but
1475 that's another story).
1477 \section pf_flexml_bypassing Bypassing the XML parser with your own C functions
1478 <b>NOTE THAT THIS DOCUMENTATION, WHILE STILL WORKING, IS STRONGLY DEPRECATED</b>
1480 So you want to bypass the XML files parser, uh? Maybe doing some parameter
1481 sweep experiments on your simulations or so? This is possible, and
1482 it's not even really difficult (well. Such a brutal idea could be
1483 harder to implement). Here is how it goes.
1485 For this, you have to first remember that the XML parsing in SimGrid is done
1486 using a tool called FleXML. Given a DTD, this gives a flex-based parser. If
1487 you want to bypass the parser, you need to provide some code mimicking what
1488 it does and replacing it in its interactions with the SURF code. So, let's
1489 have a look at these interactions.
1491 FleXML parser are close to classical SAX parsers. It means that a
1492 well-formed SimGrid platform XML file might result in the following
1495 - start "platform_description" with attribute version="2"
1496 - start "host" with attributes id="host1" power="1.0"
1498 - start "host" with attributes id="host2" power="2.0"
1500 - start "link" with ...
1502 - start "route" with ...
1503 - start "link_ctn" with ...
1506 - end "platform_description"
1508 The communication from the parser to the SURF code uses two means:
1509 Attributes get copied into some global variables, and a surf-provided
1510 function gets called by the parser for each event. For example, the event
1511 - start "host" with attributes id="host1" power="1.0"
1513 let the parser do something roughly equivalent to:
1515 strcpy(A_host_id,"host1");
1520 In SURF, we attach callbacks to the different events by initializing the
1521 pointer functions to some the right surf functions. Since there can be
1522 more than one callback attached to the same event (if more than one
1523 model is in use, for example), they are stored in a dynar. Example in
1524 workstation_ptask_L07.c:
1526 /* Adding callback functions */
1527 surf_parse_reset_parser();
1528 surfxml_add_callback(STag_surfxml_host_cb_list, &parse_cpu_init);
1529 surfxml_add_callback(STag_surfxml_prop_cb_list, &parse_properties);
1530 surfxml_add_callback(STag_surfxml_link_cb_list, &parse_link_init);
1531 surfxml_add_callback(STag_surfxml_route_cb_list, &parse_route_set_endpoints);
1532 surfxml_add_callback(ETag_surfxml_link_c_ctn_cb_list, &parse_route_elem);
1533 surfxml_add_callback(ETag_surfxml_route_cb_list, &parse_route_set_route);
1535 /* Parse the file */
1536 surf_parse_open(file);
1537 xbt_assert(!surf_parse(), "Parse error in %s", file);
1541 So, to bypass the FleXML parser, you need to write your own version of the
1542 surf_parse function, which should do the following:
1543 - Fill the A_<tag>_<attribute> variables with the wanted values
1544 - Call the corresponding STag_<tag>_fun function to simulate tag start
1545 - Call the corresponding ETag_<tag>_fun function to simulate tag end
1546 - (do the same for the next set of values, and loop)
1548 Then, tell SimGrid that you want to use your own "parser" instead of the stock one:
1550 surf_parse = surf_parse_bypass_environment;
1551 MSG_create_environment(NULL);
1552 surf_parse = surf_parse_bypass_application;
1553 MSG_launch_application(NULL);
1556 A set of macros are provided at the end of
1557 include/surf/surfxml_parse.h to ease the writing of the bypass
1558 functions. An example of this trick is distributed in the file
1559 examples/msg/masterslave/masterslave_bypass.c