Virtual Layer Research Articles

APCFS: Autonomous and Parallel Compressed File System

APCFS (Autonomous and Parallel Compressed File System) is a file system that supports fast autonomous compression at high compression rates. It is designed as a virtual layer inserted over existing file system, compressing and decompressing data by intercepting kernel calls. The system achieves enhanced compression ratio by combining two compression techniques. Speed is attained by performing the two compression techniques in parallel. Experimental results indicate good performance.

Nucleation and Growth of Oxide Layers on Stainless Steels (FeCr) Using a Virtual Oxide Layer Model

To simulate the passivation of FeCr alloys, we have developed a new model involving an explicit 3D model of the oxide layer that allows us to simulate the nucleation and the growth of the passive film. In the new model, the alloy is immersed in a virtual empty oxide lattice oriented with a given epitaxy. During the dynamic evolution, the metal cations generated by the oxidation of the alloy elements are injected into the virtual lattice where they are associated with oxygen ions coming from the solution, leading to formation of oxide nuclei, lateral growth of oxide islands, and an increase of the layer thickness. The dynamic evolution is based on the kinetic Monte Carlo (KMC) method. The KMC simulation takes into account the fundamental processes involved in the passivation mechanism: metal diffusion in the bulk and at the surface of the alloy, formation of metal cations and their injection in the oxide, nucleation and growth of the oxide layer, mass transfer through the oxide, and oxide dissolution at the oxide−solution interface. The activation energies related to the various processes are calculated using the modified embedded atom method potential or derived from experimental data. The electric field in the oxide film, considered in the new model, decreases or increases the activation energies depending on the positive or negative charge of the ions. The local value of the electric field is considered as inversely proportional to the local thickness of the oxide film. The simulations were carried out with a model of 25 Å × 25 Å × 25 Å (11 atomic planes). The results reproduce qualitatively well the experimental data. For low concentrations in Cr (<14%), the oxide nuclei do not cover completely the alloy surface and the metal is corroded. For high chromium content (>16%), the oxide layer grows, covers the whole surface, and reaches a stationary thickness of the order of 9 Å. In the intermediate zone (14−16%), the transition from incomplete or no passivation to complete passivation is continuous. The passive film is enriched with chromium. For alloys with low Cr content, extensive iron dissolution is required to obtain passivation. This leads to increased surface roughness. The oxidation process produces vacancies in the alloy that may form cavities at the oxide−metal interface or in the bulk of the alloy. For low chromium content, these cavities coalesce, leading to passivity breakdown and pit initiation.

Multicast Grooming Algorithm in Waveband Switching Optical Networks

In optical Wavelength-Division-Multiplexing (WDM) networks, multicast becomes more and more popular to provide high-speed communication between one point and multiple points. At the same time, the ports of Optical Cross-Connect (OXC) are greatly enhanced with the increasing number of wavelengths in fibers, and then the waveband switching technique is proposed to save the ports and reduce the cost of OXC. However, current waveband grooming algorithms are all limited in unicast. To achieve the multicast communication and meanwhile save the ports of OXC, we need to solve the multicast grooming, routing and wavelength/waveband assignment problem which is the HP-hard. In this paper, we propose a heuristic algorithm named Integrated Multicast Waveband Grooming (IMWG) based on Multicast Layered Auxiliary Graph (MLAG) that includes a Virtual Topology Layer (VTL) and multiple Waveband-Plane Layers (WPLs) to support the single-hop, multi-hop and hybrid multicast waveband grooming. For each demand, IMWG first computes a single-hop or multi-hop grooming waveband-tree on VTL. If the grooming waveband-tree cannot be found on VTL, IMWG computes a new waveband-tree on WPL. If the new waveband-tree cannot be found on WPL, IMWG computes a hybrid grooming waveband-tree on MLAG. Simulation results show that, compared with other algorithms, IMWG is able to obtain better performances.

DFS

We present the design, implementation, and evaluation of Direct File System (DFS) for virtualized flash storage. Instead of using traditional layers of abstraction, our layers of abstraction are designed for directly accessing flash memory devices. DFS has two main novel features. First, it lays out its files directly in a very large virtual storage address space provided by FusionIO's virtual flash storage layer. Second, it leverages the virtual flash storage layer to perform block allocations and atomic updates. As a result, DFS performs better and is much simpler than a traditional Unix file system with similar functionalities. Our microbenchmark results show that DFS can deliver 94,000 I/O operations per second (IOPS) for direct reads and 71,000 IOPS for direct writes with the virtualized flash storage layer on FusionIO's ioDrive. For direct access performance, DFS is consistently better than ext3 on the same platform, sometimes by 20%. For buffered access performance, DFS is also consistently better than ext3, and sometimes by over 149%. Our application benchmarks show that DFS outperforms ext3 by 7% to 250% while requiring less CPU power.

A Software Architecture for Adaptive Modular Sensing Systems

By combining a number of simple transducer modules, an arbitrarily complex sensing system may be produced to accommodate a wide range of applications. This work outlines a novel software architecture and knowledge representation scheme that has been developed to support this type of flexible and reconfigurable modular sensing system. Template algorithms are used to embed intelligence within each module. As modules are added or removed, the composite sensor is able to automatically determine its overall geometry and assume an appropriate collective identity. A virtual machine-based middleware layer runs on top of a real-time operating system with a pre-emptive kernel, enabling platform-independent template algorithms to be written once and run on any module, irrespective of its underlying hardware architecture. Applications that may benefit from easily reconfigurable modular sensing systems include flexible inspection, mobile robotics, surveillance, and space exploration.

NoHype

Cloud computing is a disruptive trend that is changing the way we use computers. The key underlying technology in cloud infrastructures is virtualization -- so much so that many consider virtualization to be one of the key features rather than simply an implementation detail. Unfortunately, the use of virtualization is the source of a significant security concern. Because multiple virtual machines run on the same server and since the virtualization layer plays a considerable role in the operation of a virtual machine, a malicious party has the opportunity to attack the virtualization layer. A successful attack would give the malicious party control over the all-powerful virtualization layer, potentially compromising the confidentiality and integrity of the software and data of any virtual machine. In this paper we propose removing the virtualization layer, while retaining the key features enabled by virtualization. Our NoHype architecture, named to indicate the removal of the hypervisor, addresses each of the key roles of the virtualization layer: arbitrating access to CPU, memory, and I/O devices, acting as a network device (e.g., Ethernet switch), and managing the starting and stopping of guest virtual machines. Additionally, we show that our NoHype architecture may indeed be "no hype" since nearly all of the needed features to realize the NoHype architecture are currently available as hardware extensions to processors and I/O devices.

A simplified numerical model of 3-D groundwater and solute transport at large scale area

A simplified numerical model of groundwater and solute transport is developed. At large scale area there exists a big spatial scale difference between horizontal and vertical length scales. In the resultant model, the seepage region is particularly divided into several virtual layers along the z direction and vertical 1-D columns covering x-y 2-D area according to stratum properties. The numerical algorithm is replacing the full 3-D water and mass balance analysis as the 2-D Galerkin finite element method in x- and y-directions and 1-D finite differential approach in the z direction. The reasonable method of giving minimum thickness is successfully used to handle transient change of water table, drying cells and problem of rewetting. The solution of the simplified model is preconditioned conjugate gradient and ORTHOMIN method. The validity of the developed 3-D groundwater model is tested with the typical pumping and backwater scenarios. Results of water balance of the computed example reveal the model computation reliability. Based on a representative 3-D pollution case, the solute transport module is tested against computing results using the MT3DMS. The capability and high efficiency to predict non-stationary situations of free groundwater surface and solute plume in regional scale problem is quantitatively investigated. It is shown that the proposed model is computationally effective.

Survivable routing algorithm with mixed shared protection in fault‐tolerant GMPLS multi‐layered optical networks

AbstractIn this paper, we studied survivability in general multi‐protocol label switching (GMPLS) multi‐layered optical networks and proposed a novel mixed shared protection algorithm (MSPA) based on the integrated layered graph (ILG) to tolerate single‐fibre link failure. The ILG is composed of one virtual topology layer (VTL), corresponding to IP/MPLS layer; multiple wavelength‐plane layers (WPLs), corresponding to the wavelength‐division multiplexing (WDM) optical layer; and some virtual links that connect the VTL to each WPL in order to solve the problem on routing, wavelength assignment and low‐speed traffic grooming. For every connection request, the MSPA first computes a single‐hop or multi‐hop label switching path (LSP) pair, involving a working LSP and a physical link‐disjoint backup LSP in the VTL. If the LSP pair is not found in the VTL, the MSPA computes the new LSP pair in the WPL. Conversely, if the new LSP pair is not detected in the WPL, the MSPA computes the hybrid multi‐hop LSP pair in the ILG. In the MSPA, if the two working LSPs are in physical link‐disjoint, their corresponding backup LSPs can share the backup bandwidth. In addition, the working LSPs and the backup LSPs can also share the same mixed of wavelength, in which the different bandwidths can be assigned to different working and backup LSPs. Simulation results show that, compared to previous algorithm, the MSPA can obtain a better bandwidth utilisation ratio (BUR) with a lower blocking probability. Copyright © 2010 John Wiley &amp; Sons, Ltd.

Evaluation of film adhesion to substrates by means of surface acoustic wave dispersion

For thin film structures acoustically classified as slow-on-fast systems, modeling and evaluation of their interfacial condition are known to be very complex and difficult due to dispersion and multi-mode excitation of acoustic waves. This paper presents a quantitative model and a reliable measurement procedure established for adhesion evaluation of such film structures. An effective interface model employing a virtual intermediate layer is utilized for the dispersion prediction of the surface acoustic wave, which is affected by various interfacial conditions. Through acoustic microscopy experiments, this model presents a potential method to classify the bonding condition. Comparisons with a destructive scratch test and an acoustic imaging verify the failure mode of the film structure.

An elasticity model for High Throughput Computing clusters

Different methods have been proposed to dynamically provide scientific applications with execution environments that hide the complexity of distributed infrastructures. Recently virtualization has emerged as a promising technology to provide such environments. In this work we present a generic cluster architecture that extends the classical benefits of virtual machines to the cluster level, so providing cluster consolidation, cluster partitioning and support for heterogeneous environments. Additionally the capacity of the virtual clusters can be supplemented with resources from a commercial cloud provider. The performance of this architecture has been evaluated in the execution of High Throughput Computing workloads. Results show that, in spite of the overhead induced by the virtualization and cloud layers, these virtual clusters constitute a feasible and performing HTC platform. Additionally, we propose a performance model to characterize these variable capacity (elastic) cluster environments. The model can be used to dynamically dimension the cluster using cloud resources, according to a fixed budget, or to estimate the cost of completing a given workload in a target time.

High availability using virtualization

High availability has always been one of the main problems for a data center. Till now high availability was achieved by host per host redundancy, a highly expensive method in terms of hardware and human costs. A new approach to the problem can be offered by virtualization. Using virtualization, it is possible to achieve a redundancy system for all the services running on a data center. This new approach to high availability allows the running virtual machines to be distributed over a small number of servers, by exploiting the features of the virtualization layer: start, stop and move virtual machines between physical hosts. The 3RC system is based on a finite state machine, providing the possibility to restart each virtual machine over any physical host, or reinstall it from scratch. A complete infrastructure has been developed to install operating system and middleware in a few minutes. To virtualize the main servers of a data center, a new procedure has been developed to migrate physical to virtual hosts. The whole Grid data center SNS-PISA is running at the moment in virtual environment under the high availability system.

Waveband Grooming based on Layered Auxiliary Graph in multi-domain optical networks

With the number of wavelengths on fibers keeps increasing, the size and the cost of Optical Cross-Connect (OXC) are greatly enhanced and then the control and management of optical switches become more and more complicated. Therefore, the technique called waveband switching is proposed to reduce the size and the cost of OXC; that is, to save the All-Optical (OOO) switching ports in OXC. However, the existing waveband switching algorithms are all limited in single-domain optical networks. Actually, with the scale of optical backbone keeps enlarging, the network is divided to multiple independent domains to perform the hierarchy routing for achieving the scalability. In order to reduce the size and the cost of OXC meanwhile to achieve the scalability in multi-domains, in this paper we propose a new heuristic algorithm called Waveband Grooming with Layered Auxiliary Graph (WGLAG) since the waveband grooming problem is the NP-hard to perform the inter-domain routing based on the virtual topology of multi-domain network and the intra-domain routing based on the physical topology of single-domain network. In intra-domain routing with waveband grooming of each single-domain, we propose the Layered Auxiliary Graph (LAG) that includes one virtual topology layer and multiple waveband-plane layers to compute a single-hop, or multi-hop or hybrid waveband route for each connection request based on the sub-path waveband grooming scheme. Simulation results show that, WGLAG not only can effectively save more switching ports to reduce the cost of OXC but also can obtain lower blocking probability than other algorithm.

Optimizing crash dump in virtualized environments

Crash dump, or core dump is the typical way to save memory image on system crash for future offline debugging and analysis. However, for typical server machines with likely abundant memory, the time of core dump can significantly increase the mean time to repair (MTTR) by delaying the reboot-based recovery, while not dumping the failure context for analysis would risk recurring crashes on the same problems. In this paper, we propose several optimization techniques for core dump in virtualized environments, in order to shorten the MTTR of consolidated virtual machines during crashes. First, we parallelize the process of crash dump and the process of rebooting the crashed VM, by dynamically reclaiming and allocating memory between the crashed VM and the newly spawned VM. Second, we use the virtual machine management layer to introspect the critical data structures of the crashed VM to filter out the dump of unused memory. Finally, we implement disk I/O rate control between core dump and the newly spawned VM according to user-tuned rate control policy to balance the time of crash dump and quality of services in the recovery VM. We have implemented a working prototype, Vicover, that optimizes core dump on system crash of a virtual machine in Xen, to minimize the MTTR of core dump and recovery as a whole. In our experiment on a virtualized TPC-W server, Vicover shortens the downtime caused by crash dump by around 5X.

A Vision of the Next Generation Internet: A Policy Oriented Perspective

The host centric design of the current Internet does not recognize data and end-users as integral entities of the system. The first generation of Internet has been very successful and yet business, organizations, governments are finding it difficult to enforce their policies on their networks with the same ease that they do other methods of communications and transport. Ad-Hoc solutions e.g. firewalls, NAT, middle boxes etc, that tries to mitigate these issues end up providing localized myopic fixes which often hurt the basic underlying principles of the original design. The current Internet usage is “data centric” as evidenced by the popularity of the peer-to-peer applications. Data centric view abstracts a data requestor from having to know where the data comes from. We envision the future internet to be a dynamic, heterogeneous, secure, energy efficient omnipresent network flexible enough to support innovations and policy enforcements both at the edge and the core. The first step towards the next generation is the redesign of naming and name binding mechanisms. We, therefore, propose a Policy Oriented Network Architecture (PONA) and an abstract two part protocol stack with a virtualization layer in between. PONA provides a generic architecture which allows us to implement datacentric, host-centric, and user-centric Internet architecture. We also introduce the concept of generalized communication end-points – hosts, users, data/services, instantiate the ideas with the Mapping and Negotiation layer and provide an integrated framework for the next generation Internet. Both new Internets hope to develop new, faster technologies to enhance research and communication, and it is expected that both projects will eventually improve the current commercial Internet.

Enhancing File Security by Integrating Steganography Technique in Linux Kernel

In today’s world securing file data is very important. The proposed Secure File System (SFS) provides file data security using steganographic techniques in a transparent and convenient way. The proposed SFS pushes information hiding services into the Linux kernel space, mounting it between the Virtual File System layer and underlying file system. After SFS is integrated with the Linux operating system (OS), it enables OS to provide File Data Security as its inherent functionality. SFS requires that the user creates a directory and name it with the prefix 'secrt' to store the encrypted file data, such as secrtdir. Any directory on the system with the prefix 'secrt' will basically tells the system that the newly created directory will contain secret data. All files destined to be saved on this directory will be transparently hidden in non-suspicious information on the fly without any user intervention. For hiding we use a steganographic technique in which SFS is fully compatible with all underlying storage file systems. This paper illustrates the design of SFS for Linux which extends the operating system to provide file data security as its inherent functionality. S.Saranraj BE CSE IV year 2/32, Palace Street Samathur(po), Pollachi.(tk S. Tharani BE CSE IV year 1/221, Mariamman kovil street,Somandurai (po) Pollachi(tk) The full text of the article is not available in the cache. Kindly refer the IJCA digital library at www.ijcaonline.org for the complete article. In case, you face problems while downloading the full-text, please send a mail to editor at editor@ijcaonline.org

A lightweight soft-state tracking framework for dense mobile ad hoc networks

In a mobile ad hoc network, tracking protocols need to deal with–in addition to the mobility of the target–the mobility of the intermediate nodes that maintain a track toward the target. To address this problem, we propose the MDQT (Mobility-enhanced Distributed QuadTree) tracking framework. MDQT employs a static cell abstraction to mask the mobility of the nodes and provide the illusion of a logical static network overlaid on the mobile network. MDQT implements this virtual static network layer in a lightweight/communication-free manner by exploiting the soft-state principle and the snooping feature of wireless communication. Through simulation, we study the impacts of the mobile node percentage and target mobility speed on the system performance. Simulation results show that the success rate of tracking is mostly unaffected by the speed of mobile nodes, but degrades slightly with the increases in mobile node percentage. On the other hand, the cost measurements (latency, average hops, and retry rate) are more sensitive to the mobility speed, and remain unaffected by the increase of mobile node percentage—owing to our soft-state design approach. We find that even at very high mobility speeds (50 m/s), low update rates (1 update per second), and 100% node mobility, the success rate of MDQT tracking is above 85% and the latency is comparable with that of static networks.

Modeling workloads and devices for IO load balancing in virtualized environments

Virtualization has been effective in providing performance isolation and proportional allocation of resources, such as CPU and memory between VMs by using automated distributed resource schedulers and VM migration. Storage VMotion allows users to migrate virtual hard disks from one data store to another without stopping the virtual machine. There is a dire need for an automated tool to manage storage resources more effectively by doing virtual disk placement and load balancing of workloads across multiple data stores. Applicable beyond virtualization, this problem is challenging because it requires modeling both workloads and characterizing underlying devices. Furthermore, device characteristics such as number of disks backing a LUN, disk types etc. are hidden from the hosts by the virtualization layer at the array. In this paper, we propose a storage resource scheduler (SRS) to manage virtual disk placement and automatic load balancing using Storage VMotion. Our initial results lead us to believe that we can effectively model workloads and devices to improve overall storage resource utilization in practice.

A self-configurable large-scale virtual manufacturing environment for collaborative designers

As manufacturing environments are getting distributed and increasing in size, the related virtual environments are getting larger and more closely networked together. This trend has led to a new paradigm—large-scale virtual manufacturing environment (LSVME). It supports networked and distributed virtual manufacturing to meet manufacturing system requirements. Since it contains a large number of virtual components, an effective data structure and collaborative construction methodology are needed. A metaearth architecture is proposed as the data structure for representing LSVME. This architecture consists of virtual space layer, mapping layer, library layer and ontology layers, which describe interaction among virtual components and has the ability to analyze the characteristics of virtual environment. In addition, it increases reusability of virtual components and supports self-reconfiguration for manufacturing simulation. A heuristic construction method based on graph theory is proposed using this architecture. It prevents redundant design of virtual components and contributes to an effective construction scheduling technique for collaborative designers.

Security enhanced virtual disk encryption system and technology

The authors studied and improved the transparent encryption and decryption program of virtual disk encryption software,using technology of file system filter driver and methods of auditing,preventing U disk proliferation and cache leakage. A file system filter driver layer was attached to the virtual driver layer in the program of virtual disk encryption. Test results in Truecrypt under various versions of Windows operation system show that problems of U disk proliferation and cache leakage can be avoided when the security layer is attached onto the virtual disk encryption software,and users' operations can also be monitored and audited. The results indicate that the security layer can effectively enhance security performance of the virtual disk encryption software.

Molecular dynamics and analytical Langevin equation approach for the self-diffusion constant of an anisotropic fluid

We carried out a molecular-dynamics (MD) study of the self-diffusion tensor of a Lennard-Jones-type fluid, confined in a slit pore with attractive walls. We developed Bayesian equations, which modify the virtual layer sampling method proposed by Liu, Harder, and Berne (LHB) [P. Liu, E. Harder, and B. J. Berne, J. Phys. Chem. B 108, 6595 (2004)]. Additionally, we obtained an analytical solution for the corresponding nonhomogeneous Langevin equation. The expressions found for the mean-squared displacement in the layers contain naturally a modification due to the mean force in the transverse component in terms of the anisotropic diffusion constants and mean exit time. Instead of running a time consuming dual MD-Langevin simulation dynamics, as proposed by LHB, our expression was used to fit the MD data in the entire survival time interval not only for the parallel but also for the perpendicular direction. The only fitting parameter was the diffusion constant in each layer.