[JBoss JIRA] (JGRP-2218) New payload interface
by Bela Ban (JIRA)
[ https://issues.jboss.org/browse/JGRP-2218?page=com.atlassian.jira.plugin.... ]
Bela Ban commented on JGRP-2218:
--------------------------------
{quote}
What's the use case for Payload.getInput()?
{quote}
The application has a ref to the {{Payload}} and wants to construct an object from it, via an input stream. E.g. if we have a {{CompositePayload}}, then the input stream would have to iterate through all of the individual payloads.
Re {{Bits.size()}}: yes, not very eficient as JGroups never uses UTF internally. But you can provide your own better impl
> New payload interface
> ---------------------
>
> Key: JGRP-2218
> URL: https://issues.jboss.org/browse/JGRP-2218
> Project: JGroups
> Issue Type: Feature Request
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 5.0
>
>
> h3. Goal
> Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
> h3. Reason
> Currently, having to pass a byte[] array to a message leads to unnecessary copying:
> * When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
> * When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
> * When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectPayload}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
> h3. Design
> Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
> * {{ArrayPayload}}: wraps a byte[] array with an offset and length
> * {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
> * {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
> * {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
> * {{IntPayload}}: a single integer
> * {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
> * {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
> The {{Payload}} interface has methods:
> * {{size()}}
> * {{writeTo(DataOutput)}}
> * {{readFrom(DataInput)}}
> * {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
> and possibly also
> * {{acquire()}} and
> * {{release()}} (for ref-counting)
> * {{copy()}}
> Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
> When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
> h4. Fragmentation
> When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
> h4. Reference counting
> If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
> h4. Payload management
> When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
> The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
> h4. Misc
> * Since this issue includes API changes, the version will be 5.0
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (JGRP-2218) New payload interface
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/JGRP-2218?page=com.atlassian.jira.plugin.... ]
Dan Berindei edited comment on JGRP-2218 at 9/13/17 11:58 AM:
--------------------------------------------------------------
{quote}
Yes, as Payload.getInput() returns an input stream, you have to produce the ref to the stream anyway and probably already store it in the ObjectPayload instance.
{quote}
What's the use case for {{Payload.getInput()}}?
My concern is that when {{Payload.readFrom(DataInput)}} is called, the {{DataInput}} could be a wrapper on a TCP socket's {{InputStream}}. In that case it's definitely not safe to keep a reference to the {{DataInput}} around and to read from it after {{readFrom()}} finishes.
{quote}
I was thinking of using Bits.sizeUTF() which I copied from somewhere...
{quote}
{{Bits.sizeUTF()}} touches all the characters in the string, so it's O\(n\). That's why in our marshaller implementation we write the UTF8-encoded bytes first and then [go back to write the number of bytes at the beginning|https://github.com/infinispan/infinispan/blob/f288c948c3ca33e7a...]. Although in this particular case I think we could have written the number of UTF16 code units instead of the number of bytes.
was (Author: dan.berindei):
{quote}
Yes, as Payload.getInput() returns an input stream, you have to produce the ref to the stream anyway and probably already store it in the ObjectPayload instance.
{quote}
What's the use case for {{Payload.getInput()}}?
My concern is that when {{Payload.readFrom(DataInput)}} is called, the {{DataInput}} could be a wrapper on a TCP socket's {{InputStream}}. In that case it's definitely not safe to keep a reference to the {{DataInput}} around and to read from it after {{readFrom()}} finishes.
{quote}
I was thinking of using Bits.sizeUTF() which I copied from somewhere...
{quote}
{{Bits.sizeUTF()}} touches all the characters in the string, so it's O(n). That's why in our marshaller implementation we write the UTF8-encoded bytes first and then [go back to write the number of bytes at the beginning|https://github.com/infinispan/infinispan/blob/f288c948c3ca33e7a...]. Although in this particular case I think we could have written the number of UTF16 code units instead of the number of bytes.
> New payload interface
> ---------------------
>
> Key: JGRP-2218
> URL: https://issues.jboss.org/browse/JGRP-2218
> Project: JGroups
> Issue Type: Feature Request
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 5.0
>
>
> h3. Goal
> Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
> h3. Reason
> Currently, having to pass a byte[] array to a message leads to unnecessary copying:
> * When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
> * When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
> * When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectPayload}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
> h3. Design
> Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
> * {{ArrayPayload}}: wraps a byte[] array with an offset and length
> * {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
> * {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
> * {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
> * {{IntPayload}}: a single integer
> * {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
> * {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
> The {{Payload}} interface has methods:
> * {{size()}}
> * {{writeTo(DataOutput)}}
> * {{readFrom(DataInput)}}
> * {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
> and possibly also
> * {{acquire()}} and
> * {{release()}} (for ref-counting)
> * {{copy()}}
> Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
> When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
> h4. Fragmentation
> When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
> h4. Reference counting
> If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
> h4. Payload management
> When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
> The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
> h4. Misc
> * Since this issue includes API changes, the version will be 5.0
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (JGRP-2218) New payload interface
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/JGRP-2218?page=com.atlassian.jira.plugin.... ]
Dan Berindei commented on JGRP-2218:
------------------------------------
{quote}
Yes, as Payload.getInput() returns an input stream, you have to produce the ref to the stream anyway and probably already store it in the ObjectPayload instance.
{quote}
What's the use case for {{Payload.getInput()}}?
My concern is that when {{Payload.readFrom(DataInput)}} is called, the {{DataInput}} could be a wrapper on a TCP socket's {{InputStream}}. In that case it's definitely not safe to keep a reference to the {{DataInput}} around and to read from it after {{readFrom()}} finishes.
{quote}
I was thinking of using Bits.sizeUTF() which I copied from somewhere...
{quote}
{{Bits.sizeUTF()}} touches all the characters in the string, so it's O(n). That's why in our marshaller implementation we write the UTF8-encoded bytes first and then [go back to write the number of bytes at the beginning|https://github.com/infinispan/infinispan/blob/f288c948c3ca33e7a...]. Although in this particular case I think we could have written the number of UTF16 code units instead of the number of bytes.
> New payload interface
> ---------------------
>
> Key: JGRP-2218
> URL: https://issues.jboss.org/browse/JGRP-2218
> Project: JGroups
> Issue Type: Feature Request
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 5.0
>
>
> h3. Goal
> Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
> h3. Reason
> Currently, having to pass a byte[] array to a message leads to unnecessary copying:
> * When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
> * When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
> * When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectPayload}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
> h3. Design
> Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
> * {{ArrayPayload}}: wraps a byte[] array with an offset and length
> * {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
> * {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
> * {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
> * {{IntPayload}}: a single integer
> * {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
> * {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
> The {{Payload}} interface has methods:
> * {{size()}}
> * {{writeTo(DataOutput)}}
> * {{readFrom(DataInput)}}
> * {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
> and possibly also
> * {{acquire()}} and
> * {{release()}} (for ref-counting)
> * {{copy()}}
> Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
> When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
> h4. Fragmentation
> When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
> h4. Reference counting
> If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
> h4. Payload management
> When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
> The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
> h4. Misc
> * Since this issue includes API changes, the version will be 5.0
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (JGRP-2218) New payload interface
by Bela Ban (JIRA)
[ https://issues.jboss.org/browse/JGRP-2218?page=com.atlassian.jira.plugin.... ]
Bela Ban updated JGRP-2218:
---------------------------
Description:
h3. Goal
Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
h3. Reason
Currently, having to pass a byte[] array to a message leads to unnecessary copying:
* When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
* When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
* When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectPayload}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
h3. Design
Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
* {{ArrayPayload}}: wraps a byte[] array with an offset and length
* {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
* {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
* {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
* {{IntPayload}}: a single integer
* {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
* {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
The {{Payload}} interface has methods:
* {{size()}}
* {{writeTo(DataOutput)}}
* {{readFrom(DataInput)}}
* {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
and possibly also
* {{acquire()}} and
* {{release()}} (for ref-counting)
* {{copy()}}
Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
h4. Fragmentation
When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
h4. Reference counting
If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
h4. Payload management
When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
h4. Misc
* Since this issue includes API changes, the version will be 5.0
was:
h3. Goal
Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
h3. Reason
Currently, having to pass a byte[] array to a message leads to unnecessary copying:
* When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
* When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
* When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectBuffer}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
h3. Design
Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
* {{ArrayPayload}}: wraps a byte[] array with an offset and length
* {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
* {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
* {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
* {{IntPayload}}: a single integer
* {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
* {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
The {{Payload}} interface has methods:
* {{size()}}
* {{writeTo(DataOutput)}}
* {{readFrom(DataInput)}}
* {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
and possibly also
* {{acquire()}} and
* {{release()}} (for ref-counting)
* {{copy()}}
Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
h4. Fragmentation
When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
h4. Reference counting
If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
h4. Payload management
When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
h4. Misc
* Since this issue includes API changes, the version will be 5.0
> New payload interface
> ---------------------
>
> Key: JGRP-2218
> URL: https://issues.jboss.org/browse/JGRP-2218
> Project: JGroups
> Issue Type: Feature Request
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 5.0
>
>
> h3. Goal
> Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
> h3. Reason
> Currently, having to pass a byte[] array to a message leads to unnecessary copying:
> * When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
> * When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
> * When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectPayload}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
> h3. Design
> Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
> * {{ArrayPayload}}: wraps a byte[] array with an offset and length
> * {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
> * {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
> * {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
> * {{IntPayload}}: a single integer
> * {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
> * {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
> The {{Payload}} interface has methods:
> * {{size()}}
> * {{writeTo(DataOutput)}}
> * {{readFrom(DataInput)}}
> * {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
> and possibly also
> * {{acquire()}} and
> * {{release()}} (for ref-counting)
> * {{copy()}}
> Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
> When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
> h4. Fragmentation
> When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
> h4. Reference counting
> If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
> h4. Payload management
> When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
> The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
> h4. Misc
> * Since this issue includes API changes, the version will be 5.0
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (JGRP-2218) New payload interface
by Bela Ban (JIRA)
[ https://issues.jboss.org/browse/JGRP-2218?page=com.atlassian.jira.plugin.... ]
Bela Ban commented on JGRP-2218:
--------------------------------
{quote}
If Payload.size() is an estimate that would work for us, although I have a feeling that we'd want to use our own implementation instead of ObjectPayload, and I like the sizeEstimate() name better.
{quote}
Yes, absolutely! The idea is that implementations of {{Payload}} can be registered with {{PayloadManagement}} dynamically (of course, under the same ID everywhere, and it needs to be available).
{quote}
In our Payload implementation would we have a choice to keep a reference to the DataInput and deserialize regions of it later (or never, copying it directly to our off-heap storage)?
{quote}
Yes, as {{Payload.getInput()}} returns an input stream, you have to produce the ref to the stream anyway and probably already store it in the {{ObjectPayload}} instance.
Once I have this feature ready, let's hack an example and see if this works. This _is_ an essential use case so we want to make sure we've got this covered!
{quote}
Util.size(String) seems to be the UTF-16 size, not UTF-8, and it assumes the length is a short (or maybe it's just an estimation as well).
{quote}
I was thinking of using {{Bits.sizeUTF()}} which I copied from somewhere...
> New payload interface
> ---------------------
>
> Key: JGRP-2218
> URL: https://issues.jboss.org/browse/JGRP-2218
> Project: JGroups
> Issue Type: Feature Request
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 5.0
>
>
> h3. Goal
> Change payload in {{Message}} from byte[] arrays to a {{Payload}} interface which can have multiple implementations.
> h3. Reason
> Currently, having to pass a byte[] array to a message leads to unnecessary copying:
> * When an application has a ref to an NIO (direct) {{ByteBuffer}}, the bytes in the byte buffer have to be copied into a byte[] array and then set in the message
> * When the application sends around byte[] arrays, but also wants to add some additional metadata, e.g. type (1000-byte requests/responses), it needs to create a new byte[] array of (say) 1001 bytes and copy the data (1000 bytes) plus the request type (1 byte) into the new copy. Example: {{MPerf}} and {{UPerf}}
> * When an object has to be sent (e.g. in Infinispan), the object has to be marshalled into a byte[] array (first allocation) and then added to the message. With the suggested {{ObjectBuffer}} (below), marshalling of the object would occur late, and it would be marshalled directly into the output stream of the bundler, eliminating the byte[] array allocation made by the application.
> h3. Design
> Instead of copying, the application creates an instance of {{Payload}} and sets the payload in {{Message}}. The {{Payload}} is then passed all the way down into the transport where it is marshalled and sent. There can be a number of payload implementations, e.g.
> * {{ArrayPayload}}: wraps a byte[] array with an offset and length
> * {{NioDirectPayload}}: wraps an NIO direct {{ByteBuffer}}
> * {{NioHeapPayload}}: wraps an NIO heap-based {{ByteBuffer}}
> * {{CompositePayload}}: wraps multiple Buffers. E.g. type (1 byte) and data (1000 bytes) as described above
> * {{IntPayload}}: a single integer
> * {{ObjectPayload}}: has an Object and a ClassLoader (for reading), plus a Marshaller which know how to marshal the object, this allows for objects to be passed in payloads and they're only marshalled at the end (transport).
> * {{PartialPayload}}: a ref to a {{Payload}}, with an offset and length
> The {{Payload}} interface has methods:
> * {{size()}}
> * {{writeTo(DataOutput)}}
> * {{readFrom(DataInput)}}
> * {{getInput()}}: this provides a {{DataInput}} stream for reading from the underlying payload
> and possibly also
> * {{acquire()}} and
> * {{release()}} (for ref-counting)
> * {{copy()}}
> Each payload impl has an ID and it should be possible to register new impls. A {{PayloadFactory}} maintains a mapping between IDs and impl classes.
> When marshalling a {{Payload}}, the ID is written first, followed by the payload's {{writeTo()}} method. When reading payloads, the {{PayloadFactory}} is used to create instances from IDs.
> h4. Fragmentation
> When fragmenting a buffer, the fragments are instances of {{PartialPayload}} which maintains an offset and length over an underlying payload. When marshalling a {{PartialPayload}}, only the part between offset and offset+length is written to the output stream.
> h4. Reference counting
> If we implement ref-counting, then payloads can be reused as soon as the ref-count is 0. For example, when sending a message, the payload's ref-count could be incremented by the app calling {{acquire()}}. (Assuming the message is a unicast message), {{UNICAST3}} would increment the count to 2. This is needed because {{UNICAST3}} might have to retransmit the message if it was lost on the network, and meanwhile the payload cannot be reused (changed). The app calls {{release()}} when the {{JChannel.send()}} call returns, but the payload cannot be reused until {{UNICAST3}} calls {{release()}} as well. This will happen when an {{ACK}} for the given message has been received.
> h4. Payload management
> When a request is received, the buffer is created from the bytes received on the network, based on the ID. This should be done by asking a {{PayloadManagement}} (or {{PayloadPool}} component for a new buffer. A naive implementation might create a new buffer every time, and more sophisticated one might use a pool of payloads.
> The {{PayloadManagement}} instance could be replaced by one's own implementation; this allows for an application to control the lifecycle of payloads: thus the creation of buffers by the application and of payloads received over the network can be controlled by the same payload management impl.
> h4. Misc
> * Since this issue includes API changes, the version will be 5.0
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (ELY-183) Protocols for password changing
by Jan Kalina (JIRA)
[ https://issues.jboss.org/browse/ELY-183?page=com.atlassian.jira.plugin.sy... ]
Jan Kalina commented on ELY-183:
--------------------------------
[~czacharym] I just checked the SRP protocol and it looks like it is only for authorization, not for password change - or am I looking bad?
> Protocols for password changing
> -------------------------------
>
> Key: ELY-183
> URL: https://issues.jboss.org/browse/ELY-183
> Project: WildFly Elytron
> Issue Type: Enhancement
> Components: API / SPI
> Reporter: Darran Lofthouse
> Fix For: 1.2.0.Beta5
>
>
> Potentially this is a bit of a research task, as I have mentioned in a couple of places I don't like relying on SSL exclusively for confidentiality - my reasons being it is perfect until their is a compromise and then it is as useful as a chocolate tea pot ;-)
> A lot of the emphasis in the Elytron development so far has been implementation of the more secure SASL mechanisms to eliminate weak password exchanges between a client and the server - however we still have the need for password to be set remotely, this task is to explore some of those options.
> Are there any existing protocols to remotely set a password securely?
> Is there anything specific to our current password types we can take advantage of?
> Are there features of any of our SASL mechanisms to apply a second layer of confidentiality?
> Any other options?
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months
[JBoss JIRA] (WFCORE-3289) CLI Completion issues with variables
by Jean-Francois Denise (JIRA)
Jean-Francois Denise created WFCORE-3289:
--------------------------------------------
Summary: CLI Completion issues with variables
Key: WFCORE-3289
URL: https://issues.jboss.org/browse/WFCORE-3289
Project: WildFly Core
Issue Type: Bug
Components: CLI
Reporter: Jean-Francois Denise
Assignee: Jean-Francois Denise
When variables are substituted, the command length his updated and makes completion logic to fail. The substitution needs to be taken into account at completion time.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
8 years, 9 months