Making Sensor Node Virtual Machines Work for Real-World Applications

Due to the increasing number of constituting jobs and input data size, the execution of modern complex workflow-based applications on cloud requires a large number of virtual machines (VMs), which makes the cost a great concern. Under the constraints of VM processing and storage capabilities and communication bandwidths between VMs, how to quickly figure out a cost-optimal resource provisioning and scheduling solution for a given cloud workflow is becoming a challenge. The things become even worse when taking the infrastructure-related failures with transient characteristics into account. To address this problem, this paper proposes a soft error aware VM selection and task scheduling approach that can achieve near-optimal the lowest possible cost. Under the reliability and completion time constraints by tenants, our approach can figure out a set of VMs with specific CPU and memory configurations and generate a cost-optimal schedule by allocating tasks to appropriate VMs. Comprehensive experimental results on well-known scientific workflow benchmarks show that compared with state-of-the-art methods, our approach can achieve up to 66% cost reduction while satisfying both reliability and completion time constraints.

Infrastructure as a service (IaaS) allows users to rent resources from the Cloud to meet their various computing requirements. The pay-as-you-use model, however, poses a nontrivial technical challenge to the IaaS cloud service providers: how to fast provision a large number of virtual machines (VMs) to meet users' dynamic computing requests? We address this challenge with VMThunder, a new VM provisioning tool, which downloads data blockson demandduring the VM booting process and speeds up VM image streaming by strategically integrating peer-to-peer (P2P) streaming techniques with enhanced optimization schemes such as transfer on demand, cache on read, snapshot on local, and relay on cache. In particular, VMThunder stores the original images in a share storage and in the meantime it adopts a tree-based P2P streaming scheme so that common image blocks are cached and reused across the nodes in the cluster. We implement VMThunder in CentOS Linux and thoroughly test its performance. Comprehensive experimental results show that VMThunder outperforms the state-of-the-art VM provisioning methods, with respect to scalability, latency, and VM runtime I/O performance.

Cloud computing offers new computing paradigms, capacity and flexible solutions to high performance computing (HPC) applications. For example, Hardware as a Service (HaaS) allows users to provide a large number of virtual machines (VMs) for computation-intensive applications using the HaaS model. Due to the large number of VMs and electronic components in HPC system in the cloud, any fault during the execution would result in re-running the applications, which will cost time, money and energy. In this paper we presented a proactive fault tolerance (FT) approach to HPC systems in the cloud to reduce the wall-clock execution time and dollar cost in the presence of faults. We also developed a generic FT algorithm for HPC systems in the cloud. Our algorithm does not rely on a spare node prior to prediction of a failure. We also developed a cost model for executing computation-intensive applications on HPC systems in the cloud. We analysed the dollar cost of provisioning spare nodes and checkpointing FT to assess the value of our approach. Our experimental results obtained from a real cloud execution environment show that the wall-clock execution time and cost of running computation-intensive applications in cloud can be reduced by as much as 30%. The frequency of checkpointing of computation-intensive applications can be reduced up to 50% with our FT approach for HPC in the cloud compared with current FT approaches.

From experience with wireless sensor networks it has become apparent that dynamic reprogramming of the sensor nodes is a useful feature. The resource constraints in terms of energy, memory, and processing power make sensor network reprogramming a challenging task. Many different mechanisms for reprogramming sensor nodes have been developed ranging from full image replacement to virtual machines.We have implemented an in-situ run-time dynamic linker and loader that use the standard ELF object file format. We show that run-time dynamic linking is an effective method for reprogramming even resource constrained wireless sensor nodes. To evaluate our dynamic linking mechanism we have implemented an application-specific virtual machine and a Java virtual machine and compare the energy cost of the different linking and execution models. We measure the energy consumption and execution time overhead on real hardware to quantify the energy costs for dynamic linkin.Our results suggest that while in general the overhead of a virtual machine is high, a combination of native code and virtual machine code provide good energy efficiency. Dynamic run-time linking can be used to update the native code, even in heterogeneous networks.

This paper describes how virtual classes can be supported in a vir-tual machine. Main-stream virtual machines such as the Java Vir-tual Machine and the.NET platform dominate the world today, and many languages are being executed on these virtual machines even though their embodied design choices conflict with the design choices of the virtual machine. For instance, there is a non-trivial mismatch between the main-stream virtual machines mentioned above and dynamically typed languages. One language concept that creates an even greater mismatch is virtual classes, in particular be-cause fully general support for virtual classes requires generation of new classes at run-time by mixin composition. Languages like CaesarJ and Object Teams can express virtual classes restricted to the subset that does not require run-time generation of classes, be-cause of the restrictions imposed by the Java Virtual Machine. We have chosen to support virtual classes by implementing a special-ized virtual machine, and this paper describes how this virtual ma-chine supports virtual classes with full generality.

Chapter Two - Emulation: Interpretation and Binary Translation

Idleness has to be carefully exploited in wireless sensor networks (WSNs) to save power and to accumulate the energy possibly harvested from the environment. State-of-the-art microcontroller units provide a wide range of ultra-low-power inactive modes with sub-millisecond wakeup time that can be effectively used for this purpose. At the same time they are equipped with 16-bit RISC architectures clocked at tens of MHz, which make them powerful enough to run a Java-compatible virtual machine (VM). This makes it possible to bring the benefits of a virtual runtime environment into power-constrained embedded systems. VMs, however, risk to impair the effectiveness of dynamic power management as they are seen as always-active processes by the scheduler of the operating system in spite of the idleness of the threads running on top of them. Avoiding to keep sensor nodes busy when they could be idle is mandatory for the energetic sustainability of WSNs. While most of the tasks of a sensor node are inherently event-driven, the functioning of its hardware-software components is not, so that they require to be redesigned in order to exploit idleness. This paper presents VirtualSense, an open-hardware open-source ultra-low-power reactive wireless sensor module featuring a Java-compatible VM.

Embedded systems pose unique challenges to Java application developers and virtual machine designers. Chief among these challenges is the memory footprint of both the virtual machine and the applications that run within it. With the rapidly increasing set of features provided by the Java language, virtual machine designers are often forced to build custom implementations that make various tradeoffs between the footprint of the virtual machine and the subset of the Java language and class libraries that are supported. In this paper, we present the ExoVM, a system in which an application is initialized in a fully featured virtual machine, and then the code, data, and virtual machine features necessary to execute it are packaged into a binary image. Key to this process is feature analysis, a technique for computing the reachable code and data of a Java program and its implementation inside the VM simultaneously. The ExoVM reduces the need to develop customized embedded virtual machines by reusing a single VM infrastructure and automatically eliding the implementation of unused Java features on a per-program basis. We present a constraint-based instantiation of the analysis technique, an implementation in IBM's J9 Java VM, experiments evaluating our technique for the EEMBC benchmark suite, and some discussion of the individual costs of some of Java's features. Our evaluation shows that our system can reduce the non-heap memory allocation of the virtual machine by as much as 75%. We discuss VM and language design decisions that our work shows are important in targeting embedded systems, supporting the long-term goal of a common VM infrastructure spanning from motes to large servers.

VM consolidation steps in cloud computing: A perspective review

The number of virtual machines in cloud computing is constantly changing with business requirements, which raises a great challenge for monitoring dynamic objects. Traditional static monitoring methods set fixed monitoring cycles, but it is difficult to make a suitable tradeoff between monitoring timeliness and cost in cloud computing systems with a large number of virtual machines or container instances. To address the above issues, this paper proposes an on-demand monitoring approach for cloud computing. We introduce a variable cycle mechanism and an event driven mechanism in the interaction between agents and collectors to minimize network overhead and maximize monitoring efficiency. Finally, experimental results in TPC-W benchmark show that compared to the method with a fixed monitoring cycle, our approach has a lower monitoring overhead.

Virtual Desktop Infrastructures (VDIs) are gaining popularity in cloud computing by allowing companies to deploy their office environments in a virtualized setting instead of relying on physical desktop machines. Consolidating many users into a VDI environment can significantly lower IT management expenses and enables new features such as available-anywhere desktops. However, barriers to broad adoption include the slow performance of virtualized I/O, CPU scheduling interference problems, and shared-cache contention. In this paper, we propose a new soft real-time scheduling algorithm that employs flexible priority designations (via utility functions) and automated scheduler class detection (via hypervisor monitoring of user behavior) to provide a higher quality user experience. We have implemented our scheduler within the Xen virtualization platform, and demonstrate that the overheads incurred from co-locating large numbers of virtual machines can be reduced from 66% with existing schedulers to under 2% in our system. We evaluate the benefits and overheads of using a smaller scheduling time quantum in a VDI setting, and show that the average overhead time per scheduler call is on the same order as the existing SEDF and Credit schedulers.

Virtualized cloud systems are prone to performance anomalies due to various reasons such as resource contentions, software bugs, and hardware failures. It will be a daunting task for system administrators to manually keep track of the execution status of a large number of virtual machines all the time. Anomaly prediction is an effective approach to enhancing availability and reliability of Cloud infrastructures. In this paper, we propose O-MAP, a supervised online anomaly prediction scheme based on Hidden Markov Model (HMM). Our algorithm is basically distributed and runs locally on each computing machine on the Cloud in order to achieve high scalability. Experiments performed on real data sets validate the fact that our algorithm can achieve high prediction accuracy for a range of system anomalies with low overhead to the infrastructure in Cloud.

Security at virtualization level has always been a major issue in cloud computing environment. A large number of virtual machines that are hosted on a single server by various customers/client may face serious security threats due to internal/external network attacks. In this work, we have examined and evaluated these threats and their impact on OpenStack private cloud. We have also discussed the most popular DOS (Denial-of-Service) attack on DHCP server on this private cloud platform and evaluated the vulnerabilities in an OpenStack networking component, Neutron, due to which this attack can be performed through rogue DHCP server. Finally, a solution, a game-theory based cloud architecture, that helps to detect and prevent DOS attacks in OpenStack has been proposed.

This paper proposes a real-time virtual machine (VM) provisioning framework to minimize energy consumption in a cloud data center with an efficient and novel data pre-processing, real-time clustering, instant multivariate time-series prediction and placement algorithms. The proposed framework represents an enhanced version of our recent work and it is suitable for real-time VM consolidation as it takes into consideration user behavior and VM historical usage to forecast the number of future VMs and to improve consolidation process with minimum energy consumption. The proposed real-time framework is evaluated using real data with different duration and sampling rate. The results obtained show up to 80% improvement on average.

Servers have different energy efficiency and energy proportionality (EP) due to their hardware configuration (i.e., CPU generation and memory installation) and workload. However, current virtual machine (VM) scheduling in virtualized environments will saturate servers without considering their energy efficiency and EP differences. This article will discuss EASE, the energy efficiency and proportionality aware VM scheduling approach. EASE first executes customized computing intensive, memory intensive, and hybrid benchmarks to calculate a server's energy efficiency and EP. Then it schedules VMs to servers to keep them working at their peak energy efficiency point (or optimal working range). This step improves the overall energy efficiency of the cluster and the data center. For performance guarantee, EASE migrates VMs from servers under highly contending conditions. The experimental results on real clusters show that power consumption can be saved 37.07% ~ 49.98% in the homogeneous cluster. The average completion time of the computing intensive VMs increases only 0.31 % ~ 8.49%. In the heterogeneous nodes, the power consumption of the computing intensive VMs can be reduced by 44.22 %. The job completion time can be saved by 53.80%.