Guest Editorial: Special Issue on Cloud Computing

CLOUD computing is the most recent instantiation of utility computing under service-oriented architecture (SOA). Cloud computing provides computing resources on demand and enables elastic sharing of computing resources as plug and play services to cloud service consumers, cloud partners, and cloud vendors in the cloud value chain. Users only pay for the volume and time of the resources being used, and this pay-as-you-go model attracts both business and individual users. The resource sharing in cloud computing can occur at various levels of abstraction, resulting in numerous cloud offerings, such as Infrastructure as a Service, Platform as a Service, and Software as a Service. Clouds have been made possible by building on top of services and SOA, hardware and software virtualization, web technology, and standards. Current cloud research is focused on the interactions between these underlying technologies and how to provide a high quality of service (QoS). This special issue on cloud computing originated from the very successful CLOUD 2010 conference that was held in Miami, Florida in July 2010. The theme of this conference was “Change We Are Leading,” aimed at the state-of-theart technology advances made to cloud infrastructure, and various active research areas including cloud security, cloud reliability, and cloud service discovery. Submitted papers underwent a thorough selection process where every submission was reviewed by at least three members of the program committee and only 20 percent of the papers were selected for the presentation and inclusion in the conference proceedings. After the conference, the authors of the top 30 percent of the conference papers were invited to submit an extended journal version for the IEEE Transactions on Services Computing Special Issue on Cloud Computing. The extended submission was required to have significant expansion and enhancement to its conference paper version in terms of content, scope, and quality. A second peer review process was conducted on every submitted journal paper. We selected eight high quality papers to be included in this special issue. These selected papers highlight the theme of the conference, “Change We Are Leading,” with research breakthroughs that are innovative and significant. The first paper, entitled “A Trusted Virtual Machine in an Untrusted Management Environment” by Li et al. deals with virtualization and the influence of a virtual machine (VM) on security. On one hand, virtualization is a technology in which provisioning can benefit computing systems by improving resource utilization, increasing software portability, and reliability. Moreover, it may even enhance security by providing isolated execution environments for different applications that require different levels of security. On the other hand, hypervisors could be a place to mount an attack against those environments supported by this hypervisor. Therefore, for security-critical applications, it is highly desirable to have a small trusted computing base (TCB) to minimize the surface of attacks. For many applications, it is not acceptable to trust an OS because its surface of attack can be huge. The authors of this paper propose a secure virtualization architecture that provides a secure run-time environment, network interface, and secondary storage for a guest VM. An analysis shows that the proposed architecture significantly reduces the TCB of security-critical guest VMs, which in turn improves security in an untrusted management environment. A prototype of the proposed approach using the Xen virtualization system was built, and the authors demonstrate how it can be used to facilitate secure remote computing services, while execution performance is only affected slightly. In the second paper, “VNsnap: Taking Snapshots of Virtual Networked Infrastructures in the Cloud,” the authors, Kangarlou et al., address issues in a virtual networked infrastructure (VNI) where VMs are connected by a virtual network, and from there, it leads to the realization of the concept of “Infrastructure as a Service” (IaaS). It is a critical feature that a VNI checkpoint can be used to restore the operation of the entire virtual infrastructure. The authors present VNsnap, a system that takes the distributed checkpoint of VNIs. The basic advantage of their proposed solution is that VNsnap 1) does not require any modification to the applications, libraries, or (guest) operating systems running in the VMs, and 2) only incurs seconds of overhead. When running VNsnap on top of Xen, the authors demonstrate that VNsnap is effective and efficient when executing real-world parallel and distributed applications. In the third paper, “Resource Provisioning with Budget Constraints for Adaptive Applications in Cloud Environments,” Zhu et al. claim that while making the vision of IEEE TRANSACTIONS ON SERVICES COMPUTING, VOL. 5, NO. 4, OCTOBER-DECEMBER 2012 469