A Simplified TDP with Large Tables

  • Yu ZhangEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9523)


Among the performance bottlenecks for the virtual machine, memory comes next to the I/O as the second major source of overhead to be addressed. While the SPT and TDP have proved to be quite effective and mature solutions in memory virtualization, it is not yet guaranteed that they perform equally well for arbitrary kind of workloads, especially considering that the performance of HPC workloads is more sensitive to the virtual than to the native execution environment. We propose that based on the current TDP design, modification could be made to reduce the 2D page table walk with the help of large page table. By doing this, not only the guest and host context switching due to guest page fault could be avoided, but also the second dimension of paging could be potentially simplified, which will lead to better performance.


  1. 1.
    Preiss, B.R., Eng, P.: Data Structures and Algorithms with Object-Oriented Design Patterns in Java. Wiley, Chichester (1999)Google Scholar
  2. 2.
    Hoang, G., Bae, C., Lange, J., Zhang, L., Dinda, P., Joseph, R.: A case for alternative nested paging models for virtualized systems. Comput. Archit. Lett. 9, 17–20, University of Michigan (2010)Google Scholar
  3. 3.
    Wang, X., Zang, J., Wang, Z., Luo, Y., Li, X.: Selective hardware/software memory virtualization, VEE 2011, Department of Computer Science and Technology, Beijing University, March 2011Google Scholar
  4. 4.
    Bae, C.S., Lange, J.R., Dinda, P.A.: Enhancing virtualized application performance through dynamic adaptive paging mode selection, Northwestern University and University of Pittsburgh, ICAC 2011, June 2011Google Scholar
  5. 5.
    Ahn, J., Jin, S., Huh, J.: Revisiting hardware-assisted page walks for virtualized systems. Computer Science Department, KAIST, ISCA 2012, April 2012Google Scholar
  6. 6.
    Gandhi, J., Basu, A., Hill, M.D., Swift, M.M.: Efficient memory virtualization. University of Wisconsin-Madison and AMD Research, October 2014Google Scholar
  7. 7.
    Adavanced Micro Devices Inc, AMD-V Nested Paging White Paper. Adavanced Micro Devices, July 2008Google Scholar
  8. 8.
    Bhargave, R., Serebin, B., Spadini, F., Manne, S.: Accelerating two-dimensional page walks for virtualized systems. Computing Solutions Group and Advanced Architecture & Technology Lab, March 2008Google Scholar
  9. 9.
    Barr, T.W., Cox, A.L., Rixner, S.: Translation Caching: Skip, Don’t Walk (the Page Table), Rice University, June 2010Google Scholar
  10. 10.
    Linux kernel Documentation about MMU in KVM.
  11. 11.
    Johnson, M.K.: Memory allocation. Linux Journal, issue 16, August 1995.
  12. 12.
    Rubini, A., Corbet, J.: Linux Device Drivers, 2nd edn, June 2014.

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Rechnerarchitektur, Fakultät Für InformatikTechnische Universität ChemnitzChemnitzGermany

Personalised recommendations