VirtMan: design and implementation of a fast booting system for homogeneous virtual machines in iVCE

Abstract

Internet-based virtual computing environment (iVCE) has been proposed to combine data centers and other kinds of computing resources on the Internet to provide efficient and economical services. Virtual machines (VMs) have been widely used in iVCE to isolate different users/jobs and ensure trustworthiness, but traditionally VMs require a long period of time for booting, which cannot meet the requirement of iVCE’s large-scale and highly dynamic applications. To address this problem, in this paper we design and implement VirtMan, a fast booting system for a large number of virtual machines in iVCE. VirtMan uses the Linux Small Computer System Interface (SCSI) target to remotely mount to the source image in a scalable hierarchy, and leverages the homogeneity of a set of VMs to transfer only necessary image data at runtime. We have implemented VirtMan both as a standalone system and for OpenStack. In our 100-server testbed, VirtMan boots up 1000 VMs (with a 15 GB image of Windows Server 2008) on 100 physical servers in less than 120 s, which is three orders of magnitude lower than current public clouds.

Keywords

Virtual machine / Fast booting / Homogeneity / Internet-based virtual computing environment (iVCE)

References

[1] Armbrust, M., Fox, A., Griffith, R., , 2010. A view of cloud computing. Commun. ACM, 53(4):50–58. http://dx.doi.org/10.1145/1721654.1721672

[2] Chen, Z., Zhao, Y., Miao, X., , 2009. Rapid provisioning of cloud infrastructure leveraging peer-to-peer networks. Proc. 29th IEEE Int. Conf. on Distributed Computing Systems Workshops, p.324–329.http://dx.doi.org/10.1109/ICDCSW.2009.35

[3] Flouris, M.D., Bilas, A., 2005. Violin: a framework for extensible block-level storage. Proc. 13th NASA Goddard Conf. on Mass Storage Systems and Technologies,p.128–142. http://dx.doi.org/10.1109/MSST.2005.41

[4] Flouris, M.D., Lachaize, R., Bilas, A., 2008. Orchestra: extensible block-level support for resource and data sharing in networked storage systems. Proc. 14th IEEE Int. Conf. on Parallel and Distributed Systems, p.237-244. http://dx.doi.org/10.1109/ICPADS.2008.110

[5] Krekel, H., 2015. Python Tox 2.3.1. Available from https://pypi.python.org/pypi/tox [Accessed on <Date>June 28,2015</Date>].

[6] Lagar-Cavilla, H.A., Whitney, J.A., Scannell, A.M., , 2009. SnowFlock: rapid virtual machine cloning for cloud computing. Proc. 4th ACM European Conf. on Computer systems, p.1–12.http://dx.doi.org/10.1145/1519065.1519067

[7] Lange, J.M., 2015. Python Testtools 1.8.1. Available from https://pypi.python.org/pypi/testtools [Accessed on <Date>June 28, 2015</Date>].

[8] Li, J., Li, D., Ye, Y., , 2015. Efficient multi-tenant virtual machine allocation in cloud data centers. Tsinghua Sci. Technol., 20(1):81–89. http://dx.doi.org/10.1109/TST.2015.7040517

[9] Lu, X., Wang, H., Wang, J., 2006. Internet-based virtual computing environment (iVCE): concepts and architecture. Sci. China Ser. F, 49(6):681–701. http://dx.doi.org/10.1007/s11432-006-2030-6

[10] Mao, M., Humphrey, M., 2012. A performance study on the VM startup time in the cloud. Proc. 5th Int. Conf. on Cloud Computing, p.423–430. http://dx.doi.org/10.1109/CLOUD.2012.103

[11] Meyer, D.T., Aggarwal, G., Cully, B., , 2008. Parallax: virtual disks for virtual machines. ACM SIGOPS Oper. Syst. Rev., 42(4):41–54. http://dx.doi.org/10.1145/1357010.1352598

[12] Nicolae, B., Bresnahan, J., Keahey, K., , 2011. Going back and forth: efficient multideployment and multisnapshotting on clouds. Proc. 20th Int. Symp. on High Performance Distributed Computing, p.147–158. http://dx.doi.org/10.1145/1996130.1996152

[13] Peng, C., Kim, M., Zhang, Z., , 2012. VDN: virtual machine image distribution network for cloud data centers. Proc. IEEE INFOCOM, p.181–189. http://dx.doi.org/10.1109/INFCOM.2012.6195556

[14] Razavi, K., Ion, A., Kielmann, T., 2014. Squirrel: scatter hoarding VM image contents on IaaS compute nodes. Proc. 23rd Int. Symp. on High-Performance Parallel and Distributed Computing, p.265–278. http://dx.doi.org/10.1145/2600212.2600221

[15] Shamma, M., Meyer, D.T., Wires, J., , 2011. Capo: recapitulating storage for virtual desktops. FAST, p.31–45.

[16] Smith, J.E., Nair, R., 2005. The architecture of virtual machines. Computer, 38(5):32–38. http://dx.doi.org/10.1109/MC.2005.173

[17] Wartel, R., Cass, T., Moreira, B., , 2010. Image distribution mechanisms in large scale cloud providers. Proc. 2nd Int. Conf. on Cloud Computing Technology and Science, p.112–117. http://dx.doi.org/10.1109/CloudCom.2010.73

[18] Weil, S.A., Brandt, S.A., Miller, E.L., , 2006. Ceph: a scalable, high-performance distributed file system. Proc. 7th Symp. on Operating Systems Design and Implementation, p.307–320.

[19] Zhang, Y., Liu, L., 2012. Distributed line graphs: a universal technique for designing DHTs based on arbitrary regular graphs. IEEE Trans. Knowl. Data Eng., 24(9):1556–1569. http://dx.doi.org/10.1109/TKDE.2011.258

[20] Zhang, Y., Chen, L., Lu, X., , 2010. Enabling routing control in a DHT. IEEE J. Sel. Areas Commun., 28(1):28–38. http://dx.doi.org/10.1109/JSAC.2010.100104

[21] Zhang, Y., Guo, C., Li, D., , 2015. CubicRing: enabling one-hop failure detection and recovery for distributed in-memory storage systems. Proc. 12th USENIX Symp. on Networked Systems Design and Implementation,p.529–542.

[22] Zhang, Z., Li, Z., Wu, K., , 2014. VMThunder: fast provisioning of large-scale virtual machine clusters. IEEE Trans. Parall. Distr. Syst., 25(12):3328–3338.http://dx.doi.org/10.1109/TPDS.2014.7

[23] Zhao, Y., Wu, J., Liu, C., 2014. On peer-assisted data dissemination in data center networks: analysis and implementation. Tsinghua Sci. Technol., 19(1):51–64. http://dx.doi.org/10.1109/TST.2014.6733208