System and method for zero buffer copying in a middleware environment

• US 8,856,460 B2
• Filed: 05/17/2011
• Issued: 10/07/2014
• Est. Priority Date: 09/15/2010
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Bihani, Ballav, Larsen, Staffan, Liu, Steven
• Primary Examiner(s): Li, Aimee
• Assistant Examiner(s): DO, CHRISTOPHER N
• Application Number:
  US13/109,849
• Publication Number:
  US 20120066459A1
• Time in Patent Office: 1,239 Days
• Field of Search: 714/31, 705/1, 705/26, 711/103, 711/152
• US Class Current: 711/152
• CPC Class Codes: G06F 15/17318 Parallel communications techniques, e.g. gather, scatter, reduce, roadcast, multicast, all to all
• CPC Class Codes: G06F 16/20 of structured data, e.g. relational data
• CPC Class Codes: G06F 16/2358 Change logging, detection, and notification replication G06F16/27
• CPC Class Codes: G06F 16/252 between a Database Management System and a front-end application
• CPC Class Codes: G06F 16/278 Data partitioning, e.g. horizontal or vertical partitioning
• CPC Class Codes: G06F 9/45504 Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
• CPC Class Codes: H04L 41/042 comprising distributed management centres cooperatively managing the network
• CPC Class Codes: H04L 41/0686 Additional information in the notification, e.g. enhancement of specific meta-data
• CPC Class Codes: H04L 49/358 Infiniband Switches
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1004 Server selection for load balancing
• CPC Class Codes: H04L 67/1027 Persistence of sessions during load balancing
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 67/142 Managing session states for stateless protocols; Signalling session states; State transitions; Keeping-state mechanisms
• CPC Class Codes: H04L 67/146 Markers for unambiguous identification of a particular session, e.g. session cookie or URL-encoding
• CPC Class Codes: H04L 69/14 Multichannel or multilink protocols
• CPC Class Codes: Y10T 409/308904 : Multiple row dividing head
• CPC Class Codes: Y10T 409/309072 : including cutter limited to rotary motion
• CPC Class Codes: Y10T 409/309128 : with means to adjust work support vertically
• CPC Class Codes: Y10T 409/309352 : Cutter spindle or spindle support
• CPC Class Codes: Y10T 409/309408 : with cutter holder

Abstract

Systems and methods are provided for zero buffer copying, where such a system includes one or more high performance computing systems, each including one or more processors and a high performance memory. The system further includes a user space that includes a Java virtual machine (JVM) and one or more application server instances; and a plurality of byte buffers accesible to the JVM and the one or more application server instances. When a request is received by a first application server instance data associated with the request is stored in a heap space associated with the JVM, and the JVM pins the portion of the heap space where the data is stored. The data is pushed to a first byte buffer where it is accessed by the first application server instance. A response is generated by the first application server using the data, and sent by the first application server.

Claims

1. A system for zero buffer copying, comprising:

one or more high performance computing systems, each including one or more processors and a high performance memory;

a user space on each of the one or more high performance computing systems, wherein the user space includes a Java virtual machine (JVM) and one or more application server instances, wherein each of said application server instances includes a muxer and a plurality of layers to process a request;

a plurality of byte buffers accessible to the JVM and said application server instances;

wherein when the request is received by a said application server instance

data from the request is stored on a heap space, and pushed from the heap space to a byte buffer,

wherein the byte buffer is pinned to native memory by the JVM, such that the byte buffer is prevented from being copied from a processing layer to a processing layer, and instead a same copy of the byte buffer is passed through each of the plurality of processing layers,

wherein each processing layer operates on the data in the byte buffer to participate in generating a response to the original request that supplies the data in the byte buffer,

and the response is sent out by the muxer.

2. The system of claim 1, further comprising:

a kernel space which includes support for sockets direct protocol; and

one or more byte buffer-aware streams accessible to the kernel space and the user space.

3. The system of claim 1, wherein each of the plurality of byte buffers is a Java New I/O (NIO) byte buffer.

4. The system of claim 1, wherein the request is an HTTP request.

5. The system of claim 1, wherein the portion of the heap space where the data is stored is pinned from being garbage collected.

6. The system of claim 1, wherein the plurality of processing layer include Java Server Page compiler, a servlet container, and an application server core.

7. A method for zero buffer copying, comprising:

providing one or more high performance computing systems, each including one or more processors and a high performance memory;

providing a user space on each of the one or more high performance computer systems, wherein the user space includes a Java virtual machine (JVM) and one or more application server instances, wherein each of said application server instances includes a muxer and a plurality of layers to process a request;

providing a plurality of byte buffers accessible to the JVM and the one or more application server instances;

receiving the request by a said application server instance;

storing data from the request on a heap space;

writing the data from the heap space to a byte buffer,

wherein the byte buffer is pinned to native memory by the JVM, such that the byte buffer is prevented from being copied from a processing layer to a processing layer, and instead a same copy of the byte buffer is passed through each of the plurality of processing layers,

wherein each processing laye operates on the data in the byte buffer to participate in generating a response to the original request that supplies the data in the byte buffer;

and

sending the response by the muxer in said application server instance.

8. The method of claim 7 further comprising:

providing a kernel space which includes support for sockets direct protocol; and

providing one or more byte buffer-aware streams accessible to the kernel space and the user space.

9. The method of claim 7, wherein each of the byte buffers is a Java New I/O (NIO) byte buffer.

10. The method of claim 7 wherein the request is an HTTP request.

11. The method of claim 7, wherein the portion of the heap space where the data is stored is pinned from being garbage collected.

12. The method of claim 7, wherein the plurality of processing layer include Java Server Page compiler, a servlet container, and an application server core.

13. A non-transitory computer readable storage medium storing one or more sequences of instructions for zero buffer copying, wherein said instructions, when executed by one or more processors, cause the one or more processors to execute the steps comprising:

providing a user space on each of the one or more high performance computer systems, wherein the user space includes a Java virtual machine (JVM) and one or more application server instances, wherein each of said application server instances includes a muxer and a plurality of layers to process a request;

providing a plurality of byte buffers accessible to the JVM and the one or more application server instances;

receiving the request by a said application server instance;

storing data from the request on a heap space;

writing the data from the heap space to a byte buffer,

wherein the byte buffer is pinned to native memory by the JVM, such that the byte buffer is prevented from being copied from a processing layer to a processing layer, and instead a same copy of the byte buffer is passed through each of the plurality of processing layers,

wherein each processing laye operates on the data in the byte buffer to participate in generating a response to the original request that supplies the data in the byte buffer;

and

sending the response by the muxer in said application server instance.

14. The non-transitory computer readable storage medium of claim 13, further comprising:

providing a kernel space which includes support for sockets direct protocol; and

providing one or more byte buffer-aware streams accessible to the kernel space and the user space.

15. The non-transitory computer readable storage medium of claim 13, wherein each of the byte buffers is a Java New I/O (NIO) byte buffer.

16. The non-transitory computer readable storage medium of claim 13, wherein the request is an HTTP request.

17. The non-transitory computer readable storage medium of claim 13, wherein the portion of the heap space where the data is stored is pinned from being garbage collected.

18. The non-transitory computer readable storage medium of claim 13, wherein the plurality of processing layer include Java Server Page compiler, a servlet container, and an application server core.