Physical Host Research Articles

Bolt

Cloud providers routinely schedule multiple applications per physical host to increase efficiency. The resulting interference on shared resources often leads to performance degradation and, more importantly, security vulnerabilities. Interference can leak important information ranging from a service's placement to confidential data, like private keys. We present Bolt, a practical system that accurately detects the type and characteristics of applications sharing a cloud platform based on the interference an adversary sees on shared resources. Bolt leverages online data mining techniques that only require 2-5 seconds for detection. In a multi-user study on EC2, Bolt correctly identifies the characteristics of 385 out of 436 diverse workloads. Extracting this information enables a wide spectrum of previously-impractical cloud attacks, including denial of service attacks (DoS) that increase tail latency by 140x, as well as resource freeing (RFA) and co-residency attacks. Finally, we show that while advanced isolation mechanisms, such as cache partitioning lower detection accuracy, they are insufficient to eliminate these vulnerabilities altogether. To do so, one must either disallow core sharing, or only allow it between threads of the same application, leading to significant inefficiencies and performance penalties.

Load feedback-based resource scheduling and dynamic migration-based data locality for virtual hadoop clusters in openstack-based clouds

With cloud computing technology becoming more mature, it is essential to combine the big data processing tool Hadoop with the Infrastructure as a Service (IaaS) cloud platform. In this study, we first propose a new Dynamic Hadoop Cluster on IaaS (DHCI) architecture, which includes four key modules: monitoring, scheduling, Virtual Machine (VM) management, and VM migration modules. The load of both physical hosts and VMs is collected by the monitoring module and can be used to design resource scheduling and data locality solutions. Second, we present a simple load feedback-based resource scheduling scheme. The resource allocation can be avoided on overburdened physical hosts or the strong scalability of virtual cluster can be achieved by fluctuating the number of VMs. To improve the flexibility, we adopt the separated deployment of the computation and storage VMs in the DHCI architecture, which negatively impacts the data locality. Third, we reuse the method of VM migration and propose a dynamic migration-based data locality scheme using parallel computing entropy. We migrate the computation nodes to different host(s) or rack(s) where the corresponding storage nodes are deployed to satisfy the requirement of data locality. We evaluate our solutions in a realistic scenario based on OpenStack. Substantial experimental results demonstrate the effectiveness of our solutions that contribute to balance the workload and performance improvement, even under heavy-loaded cloud system conditions.

Subscriber-Driven Interference Detection for Cloud-Based Web Services

Web services are now increasingly being hosted on public cloud infrastructure as a service platforms such as the Amazon Web service elastic compute cloud (EC2). However, previous studies have shown that the virtualized infrastructure used in public clouds can introduce contention among virtual machines (VMs) for shared physical host resources eventually leading to performance problems. Subscribers in a public cloud platform typically do not have access to metrics that can directly quantify the adverse impact of such inter-VM interference on Web service response times. We present a software probe based system to address this limitation. The probe is a lightweight application that runs on each Web service VM that needs to be monitored. We periodically measure the probe’s response time on a monitored VM. We then compare this response time with the probe’s previously recorded baseline no-interference response time when it executes in isolation on a VM of the same type. Statistically significant increase in the probe’s response time from the baseline is used to detect interference. The probe also indicates the type of contention at the physical host that causes the interference. This information can be exploited by a subscriber to mitigate the problem. Results show that our approach is quite effective over two different cloud platforms and a wide variety of workload scenarios. In particular, results indicate that Web service instances hosted on EC2 suffer from interference. Our probe was able to detect 93% of performance degradations triggered by such interference. In all these cases, the probe imposed an average overhead of only 3%–4% on the mean response time of the Web service being monitored.

Naplus: a software distributed shared memory for virtual clusters in the cloud

SummaryVirtual clusters (VCs) have exhibited various advantages over traditional cluster computing platforms by virtue of their extensibility, reconfigurability, and maintainability. As such, they have become a major execution environment for cloud‐based cluster applications. However, compared with traditional clusters, their distributed‐memory programming paradigm still remains largely unchanged, which implies that cluster applications cannot be efficiently deployed in VCs, especially when virtual machines (VMs) are running in different physical hosts. Recently, some efforts have been made to improve inter‐VM communication, resulting in many studies on how cluster applications could take advantages of VCs. However, most of them mainly focus on the situation that the VMs are all coresident on the same physical machine where the message passing mechanism is usually optimized away by exploiting the host's shared memory. In this paper, we present a design and implementation of Naplus, a kernel‐based virtual machine approach to the inter‐VM communications that are across different physical hosts. Naplus is based on Nahanni, a mechanism for shared‐memory communication in virtual environments. As such, it not only inherits the major merits of Nahanni with respect to flexible data structures and efficient synchronization but also achieves a shared‐memory paradigm among VMs. With Naplus, we enable the size of shared space to be maximized as large as the sum of each machine's local memory to accommodate cluster applications with large memory footprints. We prototype Naplus in a dual‐host system where an empirical study is conducted to show the effectiveness of the Naplus approach in achieving distributed shared memory for VCs in data centers. Copyright © 2017 John Wiley & Sons, Ltd.

Angleško-slovenski glosar virtualizacijske terminologije: usklajevanje tehničnega, terminološkega in organizacijskega dela

With the growing adoption of using virtual machines over physical hosts as a form of resource consolidation, The English-Slovene Glossary of Virtualization-related Terms encompassing management of virtual machines, cloud orchestration and data storage seemed like the next logical step.The Glossary of Virtualization-related Terms has been translated into Slovene and reviewed by experts in the fields of cloud computing, virtualization technologies and linguists. Close to 6000 terms had been localized for the Slovene market, using the advanced version of Poedit application – the editor for translating apps and websites. PoEdit automatically displays translation equivalents either from its own base (built-in translation memory) or from the base of previously translated words and phrases, which had been created and offered as opensource by other users. Based on these, it makes suggestions and, over time, learns enough to fill in frequently used strings. The translated text was then imported into its original page location – the Graphic User Interface (visible on buttons on the dashboard) of the customized ManageIQ Enterprise Virtualization Manager (EVM) software used by administrators of public and private clouds. Hence the main criterion was brevity and precision in transfering meaning across languages. This is where we encountered most problems – neologisms and existing words that acquire new meaning as a result of rapid development of virtualization technology. To avoid merely adding a suffix while the core of the word remains the same in Slovene (e.g. tenant, tenant-ov) and also to encourage further additions, comments or suggested changes the glossary has been made available on Wikipedia, the online encyclopedia.

Analysis and Evaluation of Scheduling Policies for Consolidated I/O Operations

Hypervisors' smooth operation and efficient performance has an immediate effect in the supported Cloud services. We investigate scheduling algorithms that match I/O requests originated from virtual resources, to the physical CPUs that do the actual processing. We envisage a new paradigm of virtualized resource consolidation, where I/O resources required by several Virtual Machines (VMs) in different physical hosts, are provided by one (or more) external powerful dedicated appliance(s), namely the I/O Hypervisor (IOH). For this reason I/O operations are transferred from the VMs to the IOH, where they are executed. We propose and evaluate a number of scheduling algorithms for this hypervisor model, concentrating on providing guaranteed fairness among the virtual resources. A simulator has been built that describes this model and is used for the implementation and the evaluation of the algorithms. We also analyze the performance of the different hypervisor models and highlight the importance of fair scheduling.

NEAT: Network link emulation with adaptive time dilation

In evaluating the performance of highly complex networked systems, emulation is often used as it maintains much of the realism of testbeds, while offering increased flexibility and scalability. In large emulation systems, multiple and heterogeneous virtual machines can be deployed in relatively few general purpose physical hosts. Time dilation is a technique that allows virtual time to pass at a different (and potentially variable) rate with respect to real time, allowing for increased scalability of the emulated system. In this paper we present networking links in a large emulated system employing adaptive time dilation. The link emulation focuses on accurate delay and throughput emulation while allowing varying time dilation factors. To evaluate our system, we measure the delay and throughput of the virtual links under variable system loads.

An Energy-Efficient Consolidation of Virtual Machines in Cloud Data Centers

The problem of migrating virtual machines (VMs) among different physical hosts is vital for resource utilization and carbon dioxide (CO2) minimization in cloud data centers. The main objective of this paper is to introduce the proposed technique CPU Utilization Variance (CUV). CUV is based on selecting the best VMsfrom overutilized servers and migrating them into other servers to save the utilized resources and not to violate the Service Level Agreements established between the end users and cloud service provider.CUV alsochooses the most appropriate host to allocate these VMs by calculating the minimum variance of all CPU utilization between all physical servers. CUVis implemented in a large-scale data center composed of 800 physical hosts and the results obtained by CloudSim tool are in terms of energy consumption in KWh, performance in Million Instructions per Second (MIPS), number of VM migrations and Service Level Agreement Violation. Further, a comparative study has been performed between CUV and recently researches for evaluating it, which achieved high performance and lower energy consumption without violating the service level agreements in a large-scale data centers

NDynamic Framework for Secure VM Migration over Cloud Computing

In the centralized cloud controlled environment, the decision-making and monitoring play crucial role where in the host controller (HC) manages the resources across hosts in data center (DC). HC does virtual machine (VM) and physical hosts management. The VM management includes VM creation, monitoring, and migration. If HC down,the services hosted by various hosts in DC can`t be accessed outside the DC. Decentralized VM management avoids centralized failure by considering one of the hosts from DC as HC that helps in maintaining DC in running state. Each host in DC has many VM`s with the threshold limit beyond which it can`t provide service.To maintain threshold , the host`s in DC does VM migration across various hosts. The data in migration is in the form of plaintext, the intruder can analyze packet movement and can control hosts traffic. The incorporation of security mechanism on hosts in DC helps protecting data in migration. This paper discusses an approach for dynamic HC selection, VM selection and secure VM migration over cloud environment.

Towards Trusted Location SpecificInformation for Cloud Servers

Every physical datacenter is located somewhere on the globe. It is subject to the local legislation, including data protection related laws. A cloud service can be delivered from a set of datacenters in several locations. Responsibilities of the service provider include ensuring that legal and agreed constraints are respected also by its subcontractors, for example, those providing cloud computing resources. Several countries have data protection legislation that restrict sharing copies of sensitive data to locations that do not have compliant legislation. This paper presents ideas to dependably detect location specific information, like the legislation properties, of the current physical host server executing a service.

Workload-Aware Placement of Multi-Tier Applications in Virtualized Datacenters

Virtualization as one of the leading technologies has assisted datacenters to cloudify their products and provide versatile platforms and variety of Internet services. This technology also has facilitated agile deployment of complex Internet services such as Cloud-based multi-tier applications. However implementing multi-tier applications assists providers to set up flexible and scalable services, deploying such applications in virtualized environments faces challenges, which harden collocation of disparate multi-tier applications in a shared environment. In this article, we propose a placement strategy to consolidate multi-tier applications’ tiers in a virtualized datacenter regarding workload demands of individual tiers, utilization of physical hosts’ resources and operational status of virtual machines. The placement strategy identifies over-utilized hosts based on the birth–death stochastic process and ranks tiers based on Data Envelopment Analysis-Analytic Hierarchy Process modeling to be placed on target hosts. A prioritized tier will be placed on an appropriate host though least operational interference with hosted tiers incurred. The efficiency of the proposed model is evaluated using a series of in-depth experiments. We introduce results derived from quantitative and qualitative analyses that are useful for multi-tier applications placement in virtualized datacenters. Simulation results reveal that the proposed solution excels in terms of both load distribution and energy consumption in the datacenter, while the number of unnecessary migrations and consecutive Service Level Agreement violations is considerably reduced.

Fine-grained multilayer virtualized systems analysis

With the consolidation of computer services in large cloud-based data centers, almost all applications and even application development execute in virtualized systems (VS's), sometimes nested. Whether it is inside a container, a virtual machine (VM) running on a physical host, or in a nested virtual machine, every process eventually runs on a physical CPU. Consequently, multiple virtualized systems might unknowingly compete with each other for physical resources. In this paper we study the interactions between all the VS's running on a physical machine. We introduce an analysis based on kernel tracing that erases the bounds between VS's and their host, to display a multilayer system as a single layer. As a result, it becomes possible to know exactly which process is currently running on a physical CPU, even if it is launched inside multiple layers of containers, themselves enclosed into two layers of VMs. To use this analysis, we developed in Trace Compass a view that displays a time line for each host CPU, showing across time which process is running. Moreover, the full hierarchy of the VS's is retrieved from the analysis and is displayed in the view. By using a system of dynamic and permanent filters, we added the possibility to highlight in this view either traced VMs, virtual CPUs, specific processes and containers. This last feature, combined with our view, allows to thoroughly apprehend the execution flow on the physical host, although it may involve multiple nested virtualized systems.

VMCD: A Virtual Multi-Channel Disk I/O Scheduling Method for Virtual Machines

In the era of cloud computing and big data, virtualization is gaining great popularity in storage systems. Since multiple guest virtual machines (DomUs) are running on a single physical device, disk I/O fairness among DomUs and aggregated throughput remain the challenges in virtualized environments. Although several methods have been developed for disk I/O performance virtualization among multiple DomUs, most of them suffer from one or more of the following drawbacks. (1) A fair scheduling mechanism is missing when requests converge together from multiple queues. (2) Existing methods rely on better performance of the underlying storage system such as solid state drive (SSD). (3) Throughput and latency are not considered simultaneously. To address these disadvantages, this paper presents a virtual multi-channel of disk I/O (VMCD) method that can be built on top of an ordinary storage utility, which mitigates the interference among multiple DomUs by using separated virtual channel (V-Channel) and an I/O request queue for each DomU. In our VMCD, several mechanisms are employed to enhance the I/O performance, including a credit allocation mechanism, a global monitoring strategy, and a virtual multi-channel fair scheduling algorithm. The proposed techniques are implemented on the Xen virtual disk and evaluated on Linux guest operating systems. Experiments results show that VMCD increases fairness by 70 percent approximately compared with CFQ and Anticipatory schedulers, by 30 percent approximately compared with Deadline scheduler; and enhances bandwidth utilization by 28 percent approximately compared with CFQ and Anticipatory schedulers, by 37 percent compared with Deadline in the case of three or more virtual DomUs running on the same physical host.

DFA-VMP: An efficient and secure virtual machine placement strategy under cloud environment

The problem of Virtual Machine (VM) placement is critical to the security and efficiency of the cloud infrastructure. Nowadays most research focuses on the influences caused by the deployed VM on the data center load, energy consumption, resource loss, etc. Few works consider the security and privacy issues of the tenant data on the VM. For instance, as the application of virtualization technology, the VM from different tenants may be placed on one physical host. Hence, attackers may steal secrets from other tenants by using the side-channel attack based on the shared physical resources, which will threat the data security of the tenants in the cloud computing. To address the above issues, this paper proposes an efficient and secure VM placement strategy. Firstly, we define the related security and efficiency indices in the cloud computing system. Then, we establish a multi-objective constraint optimization model for the VM placement considering the security and performance of the system, and find resolution towards this model based on the discrete firefly algorithm. The experimental results in OpenStack cloud platform indicates that the above strategy can effectively reduce the possibility of malicious tenants and targeted tenants on the same physical node, and reduce energy consumption and resource loss at the data center.

Virtual Machine Placement in Cloud Computing

Objectives: Hardware virtualization is an evolving technology, because it has the potential to facilitate the consolidation of multiple workloads on a single physical host directly through cloud. We introduce the First-Last algorithm to minimize energy consumption and migration. Methods: The problem is treated as an instance of bin packing algorithm which focuses on saving energy. However the overloading of active hosts and the resultant migration are the drawbacks of this approach. This algorithm was tested on a set of heterogeneous virtual machines which have demands in both single and multi-dimension and came either statically or dynamically to the server. Findings: The virtual machine placement problem can be treated as a bin packing problem, wherein the physical machines/hosts are the bins and the virtual machines are the objects. It can then be solved through various approaches such as first fit, best fit, next fit et cetera. Using the first fit algorithm, the virtual machine is placed into the first active host which can accommodate it. If none of the active hosts satisfy the VM’s demand, then a new host is made active. In the best fit method, all the active hosts are checked first and then the host which would suffer minimum resource wastage upon VM placement is chosen. Next fit algorithm places the VM in the last /most recently activated host. While first fit is best suited for optimal performance, best fit keeps the least number of hosts active, thereby saving energy but at the cost of performance since migration overhead increases. Next fit approach would fail unless the virtual machines have been arranged in the increasing order of their demands. The firstlast algorithm that we introduce balances on energy consumption as well as performance by reducing migration. It also reduces the time complexity to a certain extent. Applications: The algorithm has a comparatively lower time complexity than the existing virtual machine placement algorithms. It also reduces the energy consumption as well as migration to a certain extent.

Physical host factors for brown rot resistance in peach fruit

Physical host factors for brown rot resistance in peach fruit

Security Implications of Data Mining in Cloud Scheduling

Cloud providers host an increasing number of popular applications, on the premise of resource flexibility and cost efficiency . Most of these systems expose virtualized resources of different types and sizes. As instances share the same physical host to increase utilization, they contend on hardware resources, e.g., last-level cache, making them vulnerable to side-channel attacks from co-scheduled applications. In this work we show that using data mining techniques can help an adversarial user of the cloud determine the nature and characteristics of co-scheduled applications and negatively impact their performance through targeted contention injections. We design Bolt, a simple runtime that extracts the sensitivity of co-scheduled applications to various types of interference and uses this signal to determine the type of these applications by applying a set of data mining techniques. We validate the accuracy of Bolt on a 39-server cluster. Bolt correctly identifies the type and characteristics of 81 percent out of 108 victim applications, and constructs specialized contention signals that degrade their performance. We also use Bolt to find the most commonly-run applications on EC2. We hope that underlining such security vulnerabilities in modern cloud facilities will encourage cloud providers to introduce stronger resource isolation primitives in their systems.

Sizing Cleancache Allocation for Virtual Machines’ Transcendent Memory

The old virtualization idea and the new multicore technology now make it possible to consolidate multiple workloads on one physical host. This helps reduce the amount of idle resources. In particular, transcendent memory is a recent idea to gather idle memory into a pool that is shared by virtual machines (VMs). It can be viewed as a new level in the memory hierarchy, between main memory and disks. Cleancache is the part of transcendent memory that is used for caching the VMs’ clean pages. This paper shows that a Cache Miss Equation can accurately capture observed cleancache behavior, despite its unusual design. The equation can be used to dynamically size and partition cleancache for the VMs. Experiments with a variety of workloads show that this equation-based allocation can help to drastically reduce disk reads and fairly allocate cleancache space.

Comparison of VM Live Migration Process Impacts on QoS of Web Server Hosted in Cloud

Virtual Machine (VM) Live Migration improves system manageability through VMs rearrangement in their physical hosts. System managers can redistribute VMs to tackle different challenges as: VMs consolidation for energy consumption reduction, load balancing and software rejuvenation. VM live migration consists in successive transfers of memory pages from one host to another. Thus, system capacity planning may consider VM live migration overhead. This paper presents impacts of VM live migration operations on system performance of different RAM-sized VMs (512MB, 1GB and 2GB). These experiments aims to observe how RAM size changes can affect responsiveness and reliability of a Web Server hosted on a single VM. The experiment has external client which sends web requests in a constant rate through Autobench (httperf) benchmark tool. This benchmark tool provides results of Web Server errors and response time during VM live migration process. Experiments results also consider network throughput during VM live migration. The results presents that network overhead due to VM live migration causes major system performance degradation in VMs with more amount of RAM. Therefore, this paper results may be useful for managers to consider VM live migration impacts in capacity planning.

Optimal virtual network migration: A step closer for seamless resource mobility

Network Virtualization provides operators with the ability to move components of a virtual network (VN), or even the entire VN, between physical hosts, in real-time and seamlessly to the end-users.This paper addresses virtual resource mobility from a fault management perspective: (i) it proposes VN Clone migration, which requires no assumptions regarding the protocols running inside the virtual networks or its own architecture; (ii) and it proposes virtual network re-embedding-node-link formulation (VNRE-NLF), an integer linear programming formulation to solve the online virtual network re-embedding problem as a simultaneous optimization of virtual nodes and virtual links, providing the optimal bound for the migration of virtual networks from physical resources prone to failure. This approach aims at minimizing the overall VN migration cost per re-embedding: (i) the number of virtual nodes migrated; and (ii) the physical bandwidth consumption.The results are very promising: the VN Clone migration achieves no VN downtimes and it takes just a few seconds to be fully performed. This makes the VN Clone approach suitable both for non- and real-time traffic. The obtained results of VNRE-NLF show that VNs are highly resilient (i.e. resilience factor higher than 0.8) to fault events if no more than 2% of the physical resources are prone to failure. Moreover, it shows that it is not only important to have enough spare capacity to re-accommodate the virtual nodes and virtual links directly affected by the fault, but also to have additional capacity to accommodate virtual link re-assignments and virtual node migrations.