+ * Add additionnal checkings on channel values in communication [AL]
+ * New: MSG_task_get_source to see on which host a task was generated [HC]
+ * New: int MSG_task_probe_from_host(int channel, m_host_t host): returns
+ the number of tasks waiting to be received on channel and sent
+ by host. [AL]
+ * New: MSG_error_t MSG_task_get_from_host(m_task_t * task, int channel, m_host_t host);
+ waits for the first task coming from a given host.. [AL]
+
+ GRAS new functionnalities: [MQ]
+ * Enhance the parsing macro to allow the size of multidimentional objects
+ to be given thru annotations.
+ * New example (and documentation): Matrix Multiplication a la RPC
+ (as when I was young!) and fix a bunch of bugs found on the way.
+
+ GRAS performance improvements: [MQ]
+ [DataDesc]
+ * Reduce the amount of cbps creation/destruction by making it static to
+ datadesc_send/recv() and using a (newly created) cbps_reset (based on
+ dynar_reset ())
+ [Virtu]
+ * Change libdata to a set so that we can search for stuff by ID (and thus
+ reduce the insane amount of dict lookups)
+
+ [Transport]
+ * Actually implement gras_datadesc_copy() so that we don't have to mimick
+ RL communication on top of SG since it's so uneffective.
+ It may also be used for inter-thread communication in RL, one day.
+ * Use gras_datadesc_copy() to exchange messages on top of SG
+ Allows to:
+ - improve message exchange performance on top of SG
+ - deprecate transport_plugin_sg.c:gras_trp_sg_chunk_send() & recv()
+ * Don't exchange on the network the size of the used part of buffer,
+ instead, specify the possible buffer size to read().
+ Advantages:
+ - reduces the amount of read/write calls (one pair per exchange)
+ - reduces the amount of exchanged data (the size)
+ - allows to retrieve all arrived data on receiver side, if we don't need
+ it right now (subsequent read will peek the buffer)
+ - allows the receiver to proceed with the begining of the stream before
+ everything is arrived
+ - make it possible to build an iov transport (using readv/writev)
+ Extra difficulty:
+ - take care of the data with non-stable storage (like stacked data),
+ and bufferize them.
+ * If possible, TCP send uses vector I/O (when writev() is here)
+ - Don't use it for receive since we send data sizes and data on the
+ same stream, so we wouldn't be able to chain large amount of chunks
+ before having to flush the stuff to read the size.
+ * Rework the transport plugin mecanism to simplify it and reduce the
+ amount of pointer dereferencement when searching for the right function
+ to use.
+
+ * I guess that now, we do almost as few system calls as possible while
+ doing as few data copy as possible.
+
+ To improve it further, we could try to send all the sizes first and then
+ all the data (to use iov on receiving size), but it's only a partial
+ solution: when you have 2 dimensional data, the sizes of the second
+ dimension is data of the first dimension, so you need 3 streams.
+
+ I'm not sure the potential performance gains justify the coding burden.
+
+ -- Da SimGrid team <simgrid-devel@lists.gforge.inria.fr> Fri, 21 Oct 2005 14:42:20 +0200