[jboss-jira] [JBoss JIRA] (JGRP-1680) RDMA based transport

[Bela Ban (JIRA)]

     [ https://issues.jboss.org/browse/JGRP-1680?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Bela Ban updated JGRP-1680:
---------------------------

    Description: 
Investigate whether an RDMA based transport makes sense.

Advantages:
* Speed, low latency (TCP/IP is bypassed entirely)
* Low CPU usage

Disadvantages:
* JNI/C code
** Such a transport implementation would have to live outside of the JGroups repo
** Maintainability nightmare: the C code would also have to be ported to various OSes
*** Investigate Java based libs (IBM's jVerbs) and C based libs (Apache Portable Runtime?)
* High memory use, growing with cluster size: similarly to TCP, a 'group multicast' would involve N-1 sends. RDMA requires a Queue Pair (QP) for each destination. Each QP requires pinned memory (receive and send buffer), so each node would have to reserve (pin) N-1 memory buffers [1]
* High cost of RDMA adapters, NICs and wiring: only a very small fraction of users would run such a transport.


[1] http://www.hpcwire.com/hpcwire/2006-08-18/a_critique_of_rdma-1.html

  was:
Investigate whether an RDMA based transport makes sense.

Advantages:
* Speed, low latency (TCP/IP is bypassed entirely)
* Low CPU usage

Disadvantages:
* JNI/C code
** Such a transport implementation would have to live outside of the JGroups repo
** Maintainability nightmare: the C code would also have to be ported to various OSes
* High memory use, growing with cluster size: similarly to TCP, a 'group multicast' would involve N-1 sends. RDMA requires a Queue Pair (QP) for each destination. Each QP requires pinned memory (receive and send buffer), so each node would have to reserve (pin) N-1 memory buffers [1]
* High cost of RDMA adapters, NICs and wiring: only a very small fraction of users would run such a transport.


[1] http://www.hpcwire.com/hpcwire/2006-08-18/a_critique_of_rdma-1.html


    
> RDMA based transport
> --------------------
>
>                 Key: JGRP-1680
>                 URL: https://issues.jboss.org/browse/JGRP-1680
>             Project: JGroups
>          Issue Type: Feature Request
>            Reporter: Bela Ban
>            Assignee: Bela Ban
>             Fix For: 3.5
>
>
> Investigate whether an RDMA based transport makes sense.
> Advantages:
> * Speed, low latency (TCP/IP is bypassed entirely)
> * Low CPU usage
> Disadvantages:
> * JNI/C code
> ** Such a transport implementation would have to live outside of the JGroups repo
> ** Maintainability nightmare: the C code would also have to be ported to various OSes
> *** Investigate Java based libs (IBM's jVerbs) and C based libs (Apache Portable Runtime?)
> * High memory use, growing with cluster size: similarly to TCP, a 'group multicast' would involve N-1 sends. RDMA requires a Queue Pair (QP) for each destination. Each QP requires pinned memory (receive and send buffer), so each node would have to reserve (pin) N-1 memory buffers [1]
> * High cost of RDMA adapters, NICs and wiring: only a very small fraction of users would run such a transport.
> [1] http://www.hpcwire.com/hpcwire/2006-08-18/a_critique_of_rdma-1.html

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira