ACM Home Page
Please provide us with feedback. Feedback
Digital Library logoTake a look at the new version of this page: [ beta version ]. Tell us what you think.
Achieving 10 Gb/s using safe and transparent network interface virtualization
Full text PdfPdf (1.63 MB)
Source
ACM/Usenix International Conference On Virtual Execution Environments archive
Proceedings of the 2009 ACM SIGPLAN/SIGOPS international conference on Virtual execution environments table of contents
Washington, DC, USA
SESSION: Breaking barriers table of contents
Pages: 61-70  
Year of Publication: 2009
ISBN:978-1-60558-375-4
Authors
Kaushik Kumar Ram  Rice University, Houston, TX, USA
Jose Renato Santos  HP Labs, Palo Alto, CA, USA
Yoshio Turner  HP Labs, Palo Alto, CA, USA
Alan L. Cox  Rice University, Houston, TX, USA
Scott Rixner  Rice University, Houston, TX, USA
Sponsors
ACM: Association for Computing Machinery
SIGPLAN: ACM Special Interest Group on Programming Languages
SIGOPS: ACM Special Interest Group on Operating Systems
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 57,   Downloads (12 Months): 409,   Citation Count: 3
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1508293.1508303
What is a DOI?

ABSTRACT

This paper presents mechanisms and optimizations to reduce the overhead of network interface virtualization when using the driver domain I/O virtualization model. The driver domain model provides benefits such as support for legacy device drivers and fault isolation. However, the processing overheads incurred in the driver domain to achieve these benefits limit overall I/O performance. This paper demonstrates the effectiveness of two approaches to reduce driver domain overheads. First, Xen is modified to support multi-queue network interfaces to eliminate the software overheads of packet demultiplexing and copying. Second, a grant reuse mechanism is developed to reduce memory protection overheads. These mechanisms shift the bottleneck from the driver domain to the guest domains, improving scalability and enabling significantly higher data rates. This paper also presents and evaluates a series of optimizations that substantially reduce the I/O virtualization overheads in the guest domain. In combination, these mechanisms and optimizations increase the maximum throughput achieved by guest domains from 2.9Gb/s to full 10 Gigabit Ethernet link rates.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
Darren Abramson, Jeff Jackson, Sridhar Muthrasanallur, Gil Neiger, Greg Regnier, Rajesh Sankaran, Ioannis Schoinas, Rich Uhlig, Balaji Vembu, and John Wiegert. Intel virtualization technology for directed I/O. Intel Technology Journal, 10 (3), August 2006.
 
2
Advanced Micro Devices, Inc. IOMMU architectural specification. www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/34434.pdf, Feb 2007. PID 34434 Rev 1.20.
 
3
Shefali Chinni and Radhakrishna Hiremane. Virtual machine device queues. Intel Corp. White Paper, 2007.
4
 
5
 
6
Intel Corp. Intel 82598 10 GbE ethernet controller open source datasheet, 2008. Revision 2.5.
 
7
Scott Devine, Edouard Bugnion, and Mendel Rosenblum. Virtualization system including a virtual machine monitor for a computer with a segmented architecture. VMware US Patent 6397242, Oct 1998.
 
8
Keir Fraser, Steve Hand, Rolf Neugebauer, Ian Pratt, Andrew Warfield, and Mark Williams. Safe hardware access with the Xen virtual machine monitor. In 1st Workshop on Operating System and Architectural Support for the on demand IT InfraStructure (OASIS), October 2004.
 
9
Leonid Grossman. Large Receive Offload implementation in Neterion 10GbE Ethernet driver. In Ottawa Linux Symposium (OLS), 2005.
10
 
11
 
12
Kieran Mansley, Greg Law, David Riddoch, Guido Barzini, Neil Turton, and Steven Pope. Getting 10 Gb/s from Xen: Safe and fast device access from unprivileged domains. In Euro-Par 2007 Workshops: Parallel Processing, 2007.
 
13
14
 
15
 
16
Microsoft. Hyper-V architecture. http://msdn.microsoft.com/en-us/library/cc768520.aspx.
 
17
Neterion. Product brief: Neterion X3100 series. http://www.neterion.com/products/pdfs/X3100ProductBrief.pdf, 2008.
 
18
}pciiovPCI SIG. I/O virtualization. www.pcisig.com/specifications/iov/.
19
20
 
21
22
23
 
24
 
25
 
26
Edwin Zhai, Gregory D. Cummings, and Yaozu Dong. Live migration with pass-through device for linux vm. In Ottawa Linux Symposium (OLS), 2008.


Collaborative Colleagues:
Kaushik Kumar Ram: colleagues
Jose Renato Santos: colleagues
Yoshio Turner: colleagues
Alan L. Cox: colleagues
Scott Rixner: colleagues