[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16442841#comment-16442841
]
ASF subversion and git services commented on PROTON-1672:
-
Commit 663365dd715f4241848804ca1022f031765c5d98 in qpid-proton-j's branch
refs/heads/master from [~gemmellr]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=663365d ]
PROTON-1672: fix issue with sending offset and/or partial content
ReadableBuffer, plus additional arrayOffset tests
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16442651#comment-16442651
]
ASF subversion and git services commented on PROTON-1672:
-
Commit 25bc3cfbad79a750c24af39e3b3aa5740bc62cea in qpid-proton-j's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=25bc3cf ]
PROTON-1672 Fix CompositeReadableBuffer arrayOffset
The arrayOffset should return a fixed position based on the state of the
buffer (slices will normally have a non-zero value)
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16441673#comment-16441673
]
ASF subversion and git services commented on PROTON-1672:
-
Commit ea46607e776062c8555ef24a74c48a9b12bf2ca9 in qpid-proton-j's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=ea46607 ]
PROTON-1672 Fix issue when PN_TRACE_FRM is enabled
During tracing if frame is larger than max frame size an
IndexOutOfBoundsException is thrown if buffer slicing not done properly.
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16437562#comment-16437562
]
ASF subversion and git services commented on PROTON-1672:
-
Commit ef6516ea0025cb9584c3569b449bfbd9f317592e in qpid-proton-j's branch
refs/heads/master from [~gemmellr]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=ef6516e ]
PROTON-1672: disable jacaco for the python tests module, it appears it provokes
sporadic issues with jython
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16435735#comment-16435735
]
ASF subversion and git services commented on PROTON-1672:
-
Commit 1c62bc2845c336f5b061509a959b699a45344474 in qpid-proton-j's branch
refs/heads/master from [~gemmellr]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=1c62bc2 ]
PROTON-1672: replace use of Java 8+ constants in tests
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16435622#comment-16435622
]
ASF subversion and git services commented on PROTON-1672:
-
Commit 18d2cb369a6d906b2cf8403184c7e35cd00ae7c0 in qpid-proton-j's branch
refs/heads/master from [~gemmellr]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=18d2cb3 ]
PROTON-1672: replace use of Java 8+ constants
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Fix For: proton-j-0.27.0
>
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16434486#comment-16434486
]
ASF subversion and git services commented on PROTON-1672:
-
Commit ec5547152fed5a74afbe11bbb33541a2cb149fa4 in qpid-proton-j's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton-j.git;h=ec55471 ]
PROTON-1672 Handle multi-frame transfer payloads more efficiently
Replace reallocation and consolidation of transfer payloads when
multiple framed transfers are inbound. Creates a
CompositeReadableBuffer that can be used to house the assembled payload
for use in the decoder. The decoder implementation refactored to handle
ReadableBuffer as the source of bytes as well as ByteBuffer. Adds
no-copy method variants to the Sender and Receiver API such that clients
or servers can process inbound and outbound deliveries without copying
the payloads when it is known to be safe not to copy.
Adds tests and jacoco reports to validate test coverage.
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Assignee: Timothy Bish
>Priority: Major
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16388315#comment-16388315
]
Robbie Gemmell commented on PROTON-1672:
[~jr...@redhat.com], yep, this is the one I was thinking of.
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Priority: Major
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
[jira] [Commented] (PROTON-1672) Large deliveries comprising many transfers are handled inefficiently
[
https://issues.apache.org/jira/browse/PROTON-1672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16388305#comment-16388305
]
Justin Ross commented on PROTON-1672:
-
[~gemmellr], maybe this is the one to use for transfer frame rightsizing?
> Large deliveries comprising many transfers are handled inefficiently
>
>
> Key: PROTON-1672
> URL: https://issues.apache.org/jira/browse/PROTON-1672
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-j
>Affects Versions: proton-j-0.23.0
>Reporter: Keith Wall
>Priority: Major
> Attachments: JProfiler_PROTON-1672.tiff
>
>
> Running performance tests using Qpid Broker J and Qpid JMS Client (0.26.0)
> shows that receipt of large messages is very slow in comparison with the send
> of the same message. For instance, sending 300MiB bytes message takes 5
> seconds on my laptop. The receipt takes 97 seconds.
> Instrumenting the client stack shows an obvious hot-spot in Proton-J.
> {{org.apache.qpid.proton.engine.impl.TransportSession#handleTransfer}}
> re-allocates/array copies the entire delivery buffer for ever transfer that
> comprises it. This leads to a non-linear loss of performance. The stack
> include Proton-J 0.23.0, but is looks like this code is unchanged.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
