Memory Virtualization for MIPS Processor Based Cloud Server
Although the Loongson 3 processor, one representative of RISC processors based on MIPS64 instruction set, nowadays gains increasing focus and applications in Servers, memory virtualization, a critical part of a system virtual machine for such MIPS processor based Cloud Severs, meets challenges such as non-virtualizable instruction set architecture, without hardware assistant and addresses translation from guest virtual address (GVA) to host physical address (HPA) supports. However, there are few virtualization researches on MIPS processor based Cloud Server and practical virtual systems are much fewer. This paper introduces a shadow TLB based memory virtualization method. We summarize the challenges and present the key technologies such as the memory architecture, address space construction and mapping mechanism, the shadow TLB maintenance method, etc. Experimental results show that our approach can provide effective memory virtualization support for MIPS processor based Cloud Servers.
KeywordsMIPS cloud computing sever virtualization memory virtualization
Unable to display preview. Download preview PDF.
- 1.MIPS architecture, http://en.wikipedia.org/wiki/MIPS_architecture
- 2.Institute of Computing Technology Chinese Academy of Sciences: Loongson 3A processor core manual (2009)Google Scholar
- 3.Wei, X., HuiXiang, W., LiMin, X., Li, R.: KVM for MIPS. In: The 2nd International Conference on Computer and Management (accepted)Google Scholar
- 4.Edouard, B., Scott, D.: Disco: Running Commodity Operating Systems on Scalable Multiprocessors. ACM Transactions on Computer Systems 4, 143–156 (1997)Google Scholar
- 5.Virtualization, http://en.wikipedia.org/wiki/Virtualization
- 6.Adams, K., Agesen, O.: A Comparison of Software and Hardware Techniques for x86 Virtualization. VMware (2010)Google Scholar
- 7.VMware Technology: Virtualization: Architectural Considerations and Other Evaluation Criteria. VMware (2010)Google Scholar
- 8.Eyad, A., Ernie, C., et al.: Verifying shadow page table algorithms. In: Proceedings of the 2010 Conference on Formal Methods in Computer-Aided Design (FMCAD 2010), pp. 267–270. FMCAD Inc., Austin (2010)Google Scholar