Abstract
Virtualization is gaining popularity in cloud computing and has become the key enabling technology in cloud infrastructure. By replicating the virtual server state to multiple independent platforms, virtualization improves the reliability and availability of cloud systems. Unfortunately, existing Virtual Machine (VM) replication solutions were designed only for software virtualized I/O, which suffers from large performance and scalability overheads. Although hardware-assisted I/O virtualization (such as SR-IOV) can achieve close to native performance and very good scalability, they cannot be properly replicated across different physical machines due to architectural limitations (such as lack of efficient device state read/write, buffering outbound packets, etc.).
In this paper, we address those architectural limitations, by proposing ReNIC, an architectural extension to SR-IOV I/O virtualization for efficient I/O replications. We have extended Xen hypervisor and the Remus rapid checkpoint solution to support this new architectural extension. We developed a system simulator on multi-core systems to extensively evaluate ReNIC. The experimental results demonstrate that ReNIC achieves up to 54% CPU usage reduction, compared to software based I/O virtualization at runtime, and up to 16.2% performance advantage over software based I/O virtualization in rapid checkpoint. During migration, ReNIC reduces service shutdown time by about 50%, compared to device emulation and paravirtualized I/O, and over 71% compared to teaming driver.
- Adams, K. and Agesen, O. 2006. A comparison of software and hardware techniques for x86 virtualization. In Proceedings of the 12th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS'12). IEEE, 2--13. Google ScholarDigital Library
- Amazon EC2 2008. Amazon EC2 service level agreement. Amazon EC2, http://aws.amazon.com/ec2-sla/.Google Scholar
- Barham, P., Dragovic, B., Fraser, K., Hand, S., Harris, T., Ho, A., Neugebauer, R., Pratt, I., and Warfield, A. 2003. Xen and the art of virtualization. In Proceedings of the 19th ACM Symposium on Operating Systems Principles (SOSP'03). ACM, 164--177. Google ScholarDigital Library
- Ben-Yehuda, M., Day, M. D., Dubitzky, Z., Factor, M., Har'El, N., Gordon, A., Liguori, A., Wasserman, O., and Yassour, B. A. 2010. The Turtles Project: Design and Implementation of Nested Virtualization. In Proceedings of the 9th Symposium on Operating Systems Design and Implementation (OSDI'10). USENIX, 423--436. Google ScholarDigital Library
- Bradford, R., Kotsovinos, E., Feldmann, A., and Schioberg, H. 2007. Live wide-area migration of virtual machines including local persistent state. In Proceedings of the 3rd ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE'07). ACM, 169--179. Google ScholarDigital Library
- Bressoud, T. C. and Schneider, F. B. 1996. Hypervisor-based fault tolerance. ACM Trans. Computer Syst. 14, 1, 80--107. Google ScholarDigital Library
- Checconi, F., Cucinotta, T., and Stein, M. 2009. Real-time issues in live migration of virtual machines, In Proceedings of the IEEE International Conference on Parallel Processing (Euro-Par'09). IEEE, 454--466. Google ScholarDigital Library
- Clark, C., Fraser, K., Hand, S., Hansen, J. G., Jul, E., Limpach, C., Pratt, I., and Warfield, A. 2005. Live migration of virtual machines. In Proceedings of the 2nd Conference on Networked Systems Design and Implementation (NSDI'05). USENIX, 273--286. Google ScholarDigital Library
- Cucinotta, T., Anastasi, G., and Abeni, L. 2008. Real-time virtual machines. In Proceedings of the 29th IEEE Real-Time System Symposium (RTSS'08), (Work in Progress Session). IEEE.Google Scholar
- Cully, B., Lefebvre, G., Meyer, D., and Feeley, M. 2008. Remus: High availability via asynchronous virtual machine replication. In Proceedings of the 5th Conference on Networked Systems Design and Implementation (NSDI'08). USENIX, 161--171. Google ScholarDigital Library
- Dong, Y., Yang, X., Li, X., Li, J., Tian, K., and Guan, H. 2010. High performance network virtualization with SR-IOV. In Proceedings of the 16th IEEE International Symposium on High-Performance Computer Architecture (HPCA'10). IEEE, 271--280.Google Scholar
- Dunlap, G. W., King, S. T., Cinar, S., Basrai, M. A., and Chen, P. M. 2002. ReVirt: enabling intrusion analysis through virtual-machine logging and replay. In Proceedings of the 5th Symposium on Operating Systems Design and Implementation (OSDI'02). USENIX, 211--224. Google ScholarDigital Library
- Dunlap, G. W., Lucchetti, D. G., Fetterman, M. A., and Chen, P. M. 2008. Execution replay of multiprocessor virtual machines. In Proceedings of the 4th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE'08). ACM, 121--130. Google ScholarDigital Library
- Fraser, K., Hand, S., Neugebauer, R., Pratt, I., Warfield, A., and Williams, M. 2004. Safe hardware access with the Xen virtual machine monitor. In Proceedings of the 1st Workshop on Operating System and Architectural Support for the On Demand IT InfraStructure (OASIS'04).Google Scholar
- Gmach, D., Rolia, J., Cherkasova, L., Belrose, G., Turicchi, T., and Kemper, A. 2008. An integrated approach to resource pool management: policies, efficiency and quality metrics. In Proceedings of the 38th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'08). IEEE, 326--335.Google Scholar
- Guo, D., Liao, G., and Bhuyan, L. N. 2009. Performance characterization and cache-aware core scheduling in a virtualized multi-core server under 10GbE. In Proceedings of the IEEE International Symposium on Workload Characterization (IISWC'09). IEEE, 168--177. Google ScholarDigital Library
- Hines, M. R., Deshpande, U., and Gopalan, K. 2009. Post-copy live migration of virtual machines. SIGOPS Operat. Syst. Rev. 43, 3, 14--26. Google ScholarDigital Library
- Huang, W., Gao, Q., Liu, J., and Panda, D. K. 2007. High performance virtual machine migration with RDMA over modern interconnects. In Proceedings of the IEEE International Conference on Cluster Computing (CLUSTER'07). IEEE, 11--20. Google ScholarDigital Library
- Intel82598 2011. Intel® 82598 10 Gigabit Ethernet Controller. http://ark.intel.com/products/41282/Intel-82599ES-10-Gigabit-Ethernet-Controller.Google Scholar
- Intel82576 2011. Intel® 82576 Gigabit Ethernet Controller. http://ark.intel.com/products/37166/Intel-82576EB-Gigabit-Ethernet-Controller.Google Scholar
- Kadav, A. and Swift, M. M. 2009. Live migration of direct-access devices. ACM SIGOPS Operat. Syst. Rev. 43, 3, 95--104. Google ScholarDigital Library
- Liu, H., Jin, H., Liao, X., Hu, L., and Yu, C. 2009. Live migration of virtual machine based on full system trace and replay. In Proceedings of the 18th ACM International Symposium on High performance Distributed Computing (HPDC'09). ACM, 101--110. Google ScholarDigital Library
- Liao, G., Zhu, X., and Bhuyan, L. N. 2011. A new server I/O architecture for high speed networks. In Proceedings of the 17th IEEE International Symposium on High-Performance Computer Architecture (HPCA'11). IEEE, 255--265. Google ScholarDigital Library
- Liao, G., Bhuyan, L. N., Wu, W., Yu, H., and King, S. R. 2010. A new TCB cache to efficiently manage TCP sessions for Web servers. In Proceedings of the 6th ACM/IEEE Symposium on Architecture for Networking and Communication Systems (ANCS'10). ACM, 1--10. Google ScholarDigital Library
- Liu, J., Huang, W., Abali, B., and Panda, D. K. 2006. High performance VMM-bypass I/O in virtual machines. In Proceedings of the USENIX Annual Technical Conference (USENIX'06). USENIX, 3--3. Google ScholarDigital Library
- Lu, M. and Chiueh, T. 2009. Fast memory state synchronization for virtualization-based fault tolerance. In Proceedings of the 39th IEEE/IFIP International conference on Dependable Systems & Networks (DSN'09). IEEE, 534--543.Google Scholar
- Magnusson, P. S., Christensson, M., Eskilson, J., Forsgren, D., Hallberg, G., Hogberg, J., Larsson, F., Moestedt, A., and Werner, B. 2002 2002002022. Simics: A full system simulation platform. IEEE Computer. 35, 2, 50--58. Google ScholarDigital Library
- Mansley, K., Law, G., Riddoch, D., Barzini, G., Turton, N., and Pope, S. 2007. Getting 10 Gb/s from Xen: Safe and fast device access from unprivileged domains, In Proceedings of the Conference on Parallel Processing (Euro-Par'07). 224--233. Google ScholarDigital Library
- Nelson, M., Lim, B., and Hutchins, G. 2005. Fast transparent migration for virtual machines. In Proceedings of the USENIX Annual Technical Conference (USENIX'05). USENIX, 391--394. Google ScholarDigital Library
- Raj, H. and Schwan, K. 2007. High performance and scalable I/O virtualization via self-virtualized devices. In Proceedings of the 16th International Symposium on High Performance Distributed Computing (HPDC'07). ACM, 179--188. Google ScholarDigital Library
- SR-IOV 2008. PCI special interest group. http://www.pcisig.com/home.Google Scholar
- Sapuntzajis, C. P., Chandra, R., Pfaff, B., Chow, J., Lam, M. S., and Rosenblum, M. 2002. Optimizing the migration of virtual computers. In Proceedings of the 5th Symposium on Operating Systems Design and Implementation (OSDI'02). USENIX, 377--390. Google ScholarDigital Library
- Santos, J. R., Turner, Y. Janakiraman, G., and Pratt, I. 2008. Bridging the gap between software and hardware techniques for I/O virtualization, In Proceedings of the USENIX Annual Technical Conference (USENIX'08). USENIX, 29--42. Google ScholarDigital Library
- Shafer, J. and Rixner, S. 2007. RiceNIC: A reconfigurable network interface for experimental research and education. In Proceedings of the Workshop on Experimental Computer Science (ExpCS'07). ACM. Google ScholarDigital Library
- Uhlig, R., Neiger, G., Rodgers, D., Santoni, A. L., Martins, F. C. M., Anderson, A. V., Bennett, S. M., Kagi, A., Leung, F. H., and Simth, L. 2005 2005005055. Intel Virtualization Technology. IEEE Computer 38, 5, 48--56. Google ScholarDigital Library
- Vmware 2011. vSphere resource management guide. http://www.vmware.com/pdf/vsphere4/r41/vsp_41_resource_mgmt.pdf.Google Scholar
- Voorsluys, W., Broberg, J., Venugopal, S., and Buyya, R. 2009. Cost of virtual machine live migration in Clouds: A performance evaluation. In Proceedings of the 1st International Conference on Cloud Computing (CloudCom'09). IEEE, 254--265. Google ScholarDigital Library
- Watson, G., Mckeown, N., and Casado, M. 2006. NetFPGA: A tool for network research and education. In Proceedings of Workshop on Architecture Research Using FPGA Platforms (WARFP).Google Scholar
- Willmann, P., Brogioli, M., and Paj, V. S. 2004. Spinach: A libertybased simulator for programmable network interface architectures. In Proceedings of the ACM SIGPLAN/SIGBED 2004 Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES'04). ACM, 20--29. Google ScholarDigital Library
- Zec, M., Mikuc, M., and Zagar, M. 2002. Estimating the impact of interrupt coalescing delays on steady state TCP throughput. In Proceedings of the 10th International Conference on Software, Telecommunications and Computer Networks (SoftCOM'02). IEEE, Croatia, Italy.Google Scholar
- Zhai, E., Cummings, G. D., and Dong, Y. 2008. Live migration with pass-through device for linux vm. In Proceedings of Ottawa Linux Symposium (OLS'08). Vol 2, 261--268.Google Scholar
Index Terms
- ReNIC: Architectural extension to SR-IOV I/O virtualization for efficient replication
Recommendations
SRVM: Hypervisor Support for Live Migration with Passthrough SR-IOV Network Devices
VEE '16Single-Root I/O Virtualization (SR-IOV) is a specification that allows a single PCI Express (PCIe) device (ysical function or PF) to be used as multiple PCIe devices (virtual functions or VF). In a virtualization system, each VF can be directly assigned ...
SRVM: Hypervisor Support for Live Migration with Passthrough SR-IOV Network Devices
VEE '16: Proceedings of the12th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution EnvironmentsSingle-Root I/O Virtualization (SR-IOV) is a specification that allows a single PCI Express (PCIe) device (ysical function or PF) to be used as multiple PCIe devices (virtual functions or VF). In a virtualization system, each VF can be directly assigned ...
A Hypervisor Approach to Enable Live Migration with Passthrough SR-IOV Network Devices
Special TopicsSingle-Root I/O Virtualization (SR-IOV) is a specification that allows a single PCI Express (PCIe) device (physical function or PF) to be used as multiple PCIe devices (virtual functions or VF). In a virtualization system, each VF can be directly ...
Comments