Streamlined guest networking in a virtualized environment

• US 8,681,821 B1
• Filed: 09/15/2012
• Issued: 03/25/2014
• Est. Priority Date: 09/08/2009
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Vincent, Pradeep
• Primary Examiner(s): Duong, Frank
• Application Number:
  US13/620,975
• Time in Patent Office: 556 Days
• Field of Search: 370229-231, 370/235, 370/351, 370/394, 3703955-39552, 370463-477
• US Class Current: 370/474
• CPC Class Codes: H04L 69/12 Protocol engines
• CPC Class Codes: H04L 69/161 Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
• CPC Class Codes: H04L 69/166 IP fragmentation; TCP segmentation
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level

Abstract

Networking speed in a virtualized environment can be improved by utilizing an enhanced protocol and parallel stack to bypass restrictions of a conventional network stack that complies with standard networking protocols. In one embodiment, an enhanced TCP protocol enables large packets received from a guest application to bypass the guest networking stack and be received to a bottom conversion layer of the enhanced stack, which is able to segment the packet into appropriately sized segments that conform to the conventional protocol. These segments can be passed to a network interface card (NIC) or other device for encapsulation, framing, or other such processing to be transmitted to the destination as Ethernet frames or other such packets.

Claims

1. A computer-implemented method for processing network packets, comprising:

under control of one or more computer systems configured with executable instructions,

receiving a first packet to a transmit-side network layer of a first host machine, the first packet complying with a first network protocol;

determining a packet size limit for transmitting data corresponding to the first packet from the first host machine to a second host machine via a physical network interface of the first host machine;

at the transmit-side network layer, converting the first packet into one or more second packets each complying with a second network protocol and having a size of at most the packet size limit; and

sending the one or more second packets to the physical network interface, the physical network interface operable to transmit the one or more second packets to the second host machine.

2. The computer-implemented method of claim 1, wherein a first size of the first packet exceeds the packet size limit.

3. The computer-implemented method of claim 1, wherein the second network protocol corresponds to one of Transmission Control Protocol (TCP) or User Datagram Protocol (UDP).

4. The computer-implemented method of claim 1, wherein each of the second packets comprises Ethernet frames.

5. The computer-implemented method of claim 1, wherein the first packet is received at a guest operating system.

6. A computer-implemented method for processing network packets comprising:

under control of one or more computer systems configured with executable instructions,

receiving one or more first packets to a physical network interface of a host machine at a first network stack, each of the one or more first packets complying with a first network protocol;

sending the one or more first packets to a receive-side network layer of a second network stack;

at the receive-side network layer, converting the one or more first packets into a second packet complying with a second network protocol; and

transmitting the second packet through the second network stack,

wherein the first network stack is bypassed during transmission.

7. The computer-implemented method of claim 6, wherein a size of the second packet exceeds a packet size limit of the physical network interface.

8. The computer-implemented method of claim 6, wherein a guest operating system receives the second packet through the second network stack.

9. The computer-implemented method of claim 8, wherein the host machine corresponds to a hypervisor and the guest operating system executes within a virtual machine effectuated by the hypervisor.

10. The computer-implemented method of claim 6, wherein each of the one or more first packets comprises Ethernet frames, and the method further comprises:

removing framing of each of the Ethernet frames at the receive-side network layer or at the physical network interface.

11. The computer-implemented method of claim 6, wherein the second packet is transmitted through the second network stack while adhering to rules laid out by the first network protocol but violating at least one of semantics or syntax for the first network protocol.

12. The computer-implemented method of claim 6, wherein the first network protocol corresponds to one of TCP or UDP.

13. A system for processing network packets, comprising:

at least one processor;

a physical network interface; and

a memory device including instructions that, when executed by the at least one processor, cause the system to;

receive a first packet complying with a first network protocol to a transmit-side network layer of the system;

determine a packet size limit for transmitting data corresponding to the first packet from the system to a second system via the physical network interface;

at the transmit-side network layer, convert the first packet into one or more second packets each complying with a second network protocol and having a size of at most the packet size limit; and

send the one or more second packets to the physical network interface, the network interface operable to transmit the one or more second packets to the second system.

14. The system of claim 13, wherein a first size of the first packet exceeds the packet size limit.

15. The system of claim 13, wherein the second network protocol corresponds to one of TCP or UDP.

16. The system of claim 13, wherein each of the one or more second packets comprises Ethernet frames.

17. The system of claim 13, wherein the first packet is received at a guest operating system.

18. A non-transitory computer-readable storage medium including instructions for processing network packets, the instructions when executed by a processor of a computer system causing the computer system to:

receive one or more first packets to a physical network interface of a host machine at a first network stack, each of the one or more first packets complying with a first network protocol;

send the one or more first packets to a receive-side network layer of a second network stack;

at the receive-side network layer, convert the one or more first packets into a second packet complying with a second network protocol; and

transmit the second packet using the second network stack,

wherein the first network stack is bypassed during transmission.

19. The non-transitory computer-readable storage medium of claim 18, wherein a size of the second packet exceeds a size limit of the physical network interface.

20. The non-transitory computer-readable storage medium of claim 18, wherein a guest operating system receives the second packet through the second network stack.

21. The non-transitory computer-readable storage medium of claim 20, wherein the host machine corresponds to a hypervisor and the guest operating system executes within a virtual machine effectuated by the hypervisor.

22. The non-transitory computer-readable storage medium of claim 18, wherein each of the one or more first packets comprises Ethernet frames, and the instructions when executed by the processor further cause the computing system to:

remove framing of the Ethernet frames at the receive-side network layer or at the physical network interface.

23. The non-transitory computer-readable storage medium of claim 18, wherein the second packet is transmitted through the second network stack while adhering to rules laid out by the first network protocol but violating at least one of semantics or syntax for the first network protocol.

24. The non-transitory computer-readable storage medium of claim 18, wherein the first network protocol corresponds one of TCP or UDP.