Popular data concentration is a widely accepted storage energy-saving technique which places frequently-accessed data on a small subset of hard disks and spins-down other infrequently-accessed disks. Many previous studies use intuitive heuristic algorithms for data placement that promote the imbalance in the access frequencies across hard disks. However, the relevance and the optimality of such file placements have not been rigorously investigated. In this paper, we formally define the energy-saving file placement problem under the capacity and performance constraints as a combinatorial optimization problem and show the theory of the optimal file placement where the file access rates in the next period are given. Our analysis based on a stochastic process of disk state transitions gives the theoretical support for the common heuristic placement method. To examine the effectiveness of the optimal file placement, we experimentally evaluate the energy-efficiency of a test storage system using the file access rates generated from the real access traces from Flickr. The experimental results show that the energy consumption can be reduced by 31.8 percent with the optimal file placement compared to the evenly distributed file placement. We also conduct simulation experiments to confirm the energy-saving impacts in larger-scale storage systems.

Live migration allows a running operating system (OS) to be moved to another physical machine with negligible downtime. Unfortunately, live migration is not supported in bare-metal clouds, which lease physical machines rather than virtual machines to offer maximum hardware performance. Since bare-metal clouds have no virtualization software, implementing live migration is difficult. Previous studies have proposed OS-level live migration; however, to prevent user intervention and broaden OS choices, live migration should be OS-independent. In addition, the overhead of live migration mechanisms should be as low as possible. This paper introduces BLMVisor, a live migration scheme for bare-metal clouds. To achieve OS-independent and lightweight live migration, BLMVisor utilizes a very thin hypervisor that exposes physical hardware devices to the guest OS directly rather than virtualizing the devices. The hypervisor captures, transfers, and reconstructs physical device states by monitoring access from the guest OS and controlling the physical devices with effective techniques. To minimize performance degradation, the hypervisor is mostly idle after completing the live migration. A performance evaluation confirmed that the OS performance with BLMVisor is comparable to that of a bare-metal machine.

Traditional infrastructure-as-a-service (IaaS) clouds provide virtual machines as servers. However, virtualization incurs a performance overhead and prevents maximum utilization of hardware functions, so several IaaS vendors have started new services called bare-metal clouds that provide physical rather than virtual machines, allowing users to have direct access to physical hardware in the cloud. Unfortunately, exposing physical hardware to users causes a hardware protection issue for cloud vendors. Since physical hardware uses non-volatile memory (NVM) to store firmware code and configuration data, this is also exposed to users. If the NVM is modified by malicious users, the hardware could be permanently corrupted or infected by malware without being noticed. This is difficult for cloud vendors to prevent because bare-metal clouds have no virtualization layer to protect their hardware. In this paper, we describe the types of attacks that are possible for bare-metal clouds and propose BMCArmor, a hardware protection scheme for baremetal clouds. BMCArmor uses a thin hypervisor that does not virtualize the hardware, just preventing access to NVM. Our experiments show that BMCArmor can successfully protect hardware while incurring little performance overhead.

Prompt acknowledgment (ACK) of received packets is key to achieving high throughputs in the Transmission Control Protocol (TCP) data flows. This also applies to Multipath TCP (MPTCP), a variant of TCP that bundles multiple paths into a single logical flow. However, the original MPTCP takes advantage of multiple paths only when sending data packets, not ACK packets. In the original MPTCP, an ACK packet goes through the same path that its originating packet has come through, even if a better path is available. The path that the originating packet came through is not always an appropriate path because the characteristics of the path (RTT, bandwidth, etc.) can be diverse. However, it is possible to improve the promptness of ACKing if we can select the appropriate path that an ACK packet goes through, regardless of what path the original data packets came through. In this study, we propose a method for improving MPTCP throughputs by improving the promptness of receiving ACK. The method is referred to as HayACK. HayACK returns an ACK packet via a path with the lowest round-trip time (RTT), even if it is not a path that received a data packet. To examine the effectiveness of our proposal, we conducted experiments in Linux using our modified version of a MPTCP protocol stack. The evaluation results show that HayACK can improve MPTCP throughputs by a maximum of 1.74x, and by 1.52x on average.

In order to control risks and facilitate effective decision-making, the relations among risk chains should be systematically analyzed, which is a very difficult process. The aim of this research is to understand the connective generating structure of risk chains and plan problem solving accordingly. Therefore, in order to select the analysis method, association rule mining was applied. From the extracted data, the height of the sources of a risk chain and recurrence nature of a risk could be discovered. It is expected that these results can prevent the occurrence of risk chains caused by human factors.

Bare-metal clouds are an emerging infrastructure-as-a-service (IaaS) that leases physical machines (bare-metal instances) rather than virtual machines, allowing resource-intensive applications to have exclusive access to physical hardware. Unfortunately, bare-metal instances require time-consuming or OS-specific tasks for deployment due to the lack of virtualization layers, thereby sacrificing several beneficial features of traditional IaaS clouds such as agility, elasticity, and OS transparency. We present BMcast, an OS deployment system with a special-purpose de-virtualizable virtual machine monitor (VMM) that supports quick and OS-transparent startup of bare-metal instances. BMcast performs streaming OS deployment while allowing direct access to physical hardware from the guest OS, and then disappears after completing the deployment. Quick startup of instances improves agility and elasticity significantly, and OS transparency greatly simplifies management tasks for cloud customers. Experimental results have confirmed that BMcast initiated a bare-metal instance 8.6 times faster than image copying, and database performance on BMcast during streaming OS deployment was comparable to that on a state-of-the-art VMM without performing deployment. BMcast incurred zero overhead after de-virtualization.

We present an optimistic primary-backup (so-called passive replication) mechanism for highly available Internet services on intercloud platforms. Our proposed method aims at providing Internet services despite the occurrence of a large-scale disaster. To this end, each service in our method creates replicas in different data centers and coordinates them with an optimistic consensus algorithm instead of a majority-based consensus algorithm such as Paxos. Although our method allows temporary inconsistencies among replicas, it eventually converges on the desired state without an interruption in services. In particular, the method tolerates simultaneous failure of the majority of nodes and a partitioning of the network. Moreover, through interservice communications, members of the service groups are autonomously reorganized according to the type of failure and changes in system load. This enables both load balancing and power savings, as well as provisioning for the next disaster. We demonstrate the service availability provided by our approach for simulated failure patterns and its adaptation to changes in workload for load balancing and power savings by experiments with a prototype implementation.

We present a game-theoretic approach for power reduction in large-scale distributed storage systems. The key idea is to use distributed hash tables to dynamically migrate virtual nodes, thus skewing the workload towards a subset of physical disks without overloading them. To realize this idea in an autonomous way (i.e., without any kind of central controller), virtual nodes are considered to be selfish agents playing a game in which each node receives a payoff according to the workload of the disk on which it currently resides. We model this setting as a potential game, where an increase in the payoff to a virtual node reduces the power of the system. This game consists of a pair of global and private utility functions, derived by means of the Wonderful Life Utility technique. The former function evaluates the state of the system, and the latter provides criteria for the migration of each node. The performance of our method is measured by simulations and a prototype implementation. From these evaluations, we find that our method reduces the running time of the disks in active mode by 12.7-18.7%, with an overall average response time of 50-190 ms.

A Life Log Service that handles action records, such as an individual search logs and data on shoppers' browsing and buying habits, have attracted attention with the spread of the Internet. Life Log Service is thought to present various risks to private information, such as personal information. For this reason, countermeasures to these risks need to be investigated. We have already qualitatively analyzed the risks of life log services. To extract the specific risks of using a life log service comprehensively, we used a Risk Breakdown Structure (RBS), which is the typical risk-analysis method. Furthermore, after risks were analyzed, concrete countermeasures were proposed. However, these results need to be quantitatively evaluated from a more practical viewpoint. In this paper, the validity is visualized by performing quantitive risk assessment on the proposed countermeasures for risks in the life log obtained in our previous research. Specifically, the risk value based on a risk formula is computed to a risk factor and its proposed countermeasures. Thereby, the effects of the proposed countermeasures on risks of the life log service found in previous research are evaluated quantitatively, and this will contribute to spreading and promoting the life log service.

We present a power-saving method for large-scale distributed storage systems. The key idea is to use virtual nodes and migrate them dynamically so as to skew the workload towards a small number of disks while not overloading them. Our proposed method consists of two kinds of algorithms, one for gathering or spreading virtual nodes according to the daily variation of workloads so that the active disks are reduced to a minimum, the other for coping with the changes in the popularity of data over a longer period. For this dynamic migration, data stored in virtual nodes are managed by a distributed hash table. Furthermore, to improve the reliability as well as to reduce the migration cost, we also propose an extension of our method by introducing a replication mechanism. The performance of our method is measured both by simulation and a prototype implementation. From the experiments, we observed that our method skews the workload so that the average load for the active physical nodes as a function of the overall capacity is 67%. At the same time, we maintain a preferred response time by setting a suitable maximum workload for each physical node.

We propose HyperShield, which is a hypervisor that can be inserted into and removed from a running operating system, for improving security. While many existing security-oriented hypervisors require modifying or rebooting an overlying operating system, HyperShield does not require this. HyperShield is intended to be a general framework for various security mechanisms. The current implementation provides two mechanisms for preventing kernel-level buffer overflow. One detects the execution of user code with the kernel privilege, and the other detects malicious modification of a return address in a control stack. HyperShield is implemented on Linux as a loadable kernel module. When the module is inserted, it places itself under the operating system and executes as a hypervisor. The operating system is migrated into a virtual machine and managed by the hypervisor. HyperShield detects attacks by combining virtualization of memory management with a hardware-assisted execution-bit feature. We have confirmed through experiments that HyperShield successfully prevented kernel-level buffer overflow attacks.

We have designed and implemented a scripting environment called "Kumoi" for managing collective VMs in a large-scale data center. Kumoi is unlike other scripting environments because it exploits strong typing with type inference and high-level description. Kumoi introduces several advancements, including treating virtual machines as first-class objects and decoupling the scripting model and its execution for hiding as many details as possible. We implemented Kumoi as an embedded domain-specific language based on Scala along with distributed agents running on each physical machine. Evaluation using example scripts showed that an administrator can more concisely write the instructions for performing complex VM lifecycle management tasks. Use of this environment should improve management efficiency and agility.

Virtual machine monitors (VMMs), including hypervisors, are a popular platform for implementing various security functionalities. However, traditional VMMs require numerous components for providing virtual hardware devices and for sharing and protecting system resources among virtual machines (VMs), enlarging the code size of and reducing the reliability of the VMMs.This paper introduces a hypervisor architecture, called parapass-through, designed to minimize the code size of hypervisors by allowing most of the I/O access from the guest operating system (OS) to pass-through the hypervisor, while the minimum access necessary to implement security functionalities is completely mediated by the hypervisor. This architecture uses device drivers of the guest OS to handle devices, thereby reducing the size of components in the hypervisor to provide virtual devices. This architecture also allows to run only single VM on it, eliminating the components for sharing and protecting system resources among VMs.We implemented a hypervisor called BitVisor and a parapass-through driver for enforcing storage encryption of ATA devices based on the parapass-through architecture. The experimental result reveals that the hypervisor and ATA driver require approximately 20 kilo lines of code (KLOC) and 1.4 KLOC respectively.

In recent years, as the data processed by governmental or commercial organizations increases, cases involving information leak have risen. It is difficult to control information on many distributed end-point computers using conventional security mechanisms. Therefore, we have been proposed a novel secure VMM (Virtual Machine Monitor) architecture which is used as a foundation of security policy enforcement on distributed computers. This paper especially introduces Role-based Access Control (RBAC) to theID management framework in a secure VMM system. Our proposal will reduce costs for distributed policies updates. Proposed RBAC mechanism employs attribute certificates (ACs) to handle user’s roles. This paper shows design and prototype implementation based on PKI-based ID card and proven open source VMM software, QEMU.

This paper presents a sketch-based password authentication system called Scribble-a-Secret as a graphical password scheme in which free-form drawings are used as a means to authenticate users. Unlike existing schemes, this approach requires no input of graphical passwords in particular sequences of strokes. Moreover, the system allows for a modicum of variation when users recreate their passwords. Our technique uses edge orientations extracted from sketch images to discern one user from another. Our experiments show that our recognition technique is robust for recognizing sketches while differentiating from others with both a false acceptance rate and false rejection rate of less than 1%.

This paper presents an approach to enhancing B-tree indexing performance by using a replication technique called persistent caching. A notable feature of the approach is its compatibility with ordinary B-trees; it exploits only the otherwise unused area of each B-tree page, and the basic behavior of B-trees need not be changed. This paper evaluates the performance of persistently cached B-trees by showing the result of mathematical analysis and of experimental investigations.

This paper describes a framework for developing mobile software robots by using the PLANET mobile object system, which is characterized by a language-neutral layered architecture, the native code execution of mobile objects, and asynchronous object passing. We propose an approach to implementing mobile Web search robots that takes full advantage of these characteristics, and we base our discussion of its effectiveness on experiments conducted in the Internet environment. The results show that the PLANET approach to mobile Web search robots significantly reduces the amount of data transferred via the Internet and that it enables the robots to work more efficiently than the robots in the conventional stationary scheme whenever nontrivial amounts of HTML files are processed.

Remote procedure call (RPC) systems have been proven to be a practical basis for building distributed applications. The RPC technique abstracts a typical communication pattern to an ordinary procedure call. Compared with an ordinary procedure call, however, the conventional RPC technique has one evident restriction; pointers (addresses) cannot be passed to remote procedures without the explicit and nontrivial programming effort. This paper presents a method that eliminates this restriction. The method enables transparent treatment of pointers in RPC by combining three key techniques: virtual memory manipulation, pointer swizzling, and coherency protocol. The experiments performed using an implementation of the method show that the method provides performance that is scalable to the access ratio of the remotely referenced data.>

The authors propose an information management system providing distribution and persistency. By separating context from virtual address space, the system has a unified approach for both distribution and persistency. The former is achieved by moving contents between sites and the latter by moving contents between virtual address space and persistent storage. Contents include any information including data, program, and even the state of execution of a program. Contents are stored persistently in a logical space termed the distributed shared repository (DSR). A programming model for the DSR is proposed. Using the model, persistency, fine-grain mobility of information, and the passing of various distributed parameters can be obtained. The implementation anti experimental performance of the system are also presented.>

We propose a method to extend an ML-style polymorphic language with transparent communication primitives, and give their precise operational semantics. These primitives allow any polymorphic programs definable in ML to be used remotely in a manner completely transparent to the programmer. Furthermore, communicating programs may be based on different architecture and use different data representations.

Multiprotocol tuple space (MTS), a distributed implementation of tuple space, is discussed. Although tuple space provides primitives convenient for describing several types of communication and data sharing, its naive implementation in distributed environments is not efficient without special hardware. MTS is a replication-based implementation on conventional workstations and a network. It provides several different replication-maintenance protocols that exploit weak consistency. The selection of protocols to achieve efficient replication maintenance is based on communication patterns.>

Many recent complex object database systems support the concepts of object identity and object identifier. Following an object identifier to access the referenced object is called navigation operation and is an essential operation in dealing with complex objects. Navigation operation is a difficult operation to implement efficiently since every navigation operation inherently causes one disk access operation. A scheme to notably accelerate the navigation operation among a sea of complex objects, by increasing the effective number of objects in one disk page is proposed. The main concept of the presented technique is threefold. The first idea is to store a cached value within a complex object that is referencing another complex object. The second is that when the referenced object is to be updated the update propagation is delayed until the time when the cached value is referenced. The third is to utilize a hashed table on main memory to efficiently validate the consistency between the cached values and the original values.>

The mechanisms for executing concurrent applications proposed so far fall into one of three groups: processes, kernel-level threads, and user-level threads. Each of them is insufficient in terms of either parallelism, the flexibility to combine separately developed programs at run-time, or costs of operations such as creation, switching, and destruction. A thread facility in the XERO operating system overcomes this problem and provides a uniform framework for executing concurrent applications. To achieve parallelism of threads, the flexibility to combine separately developed programs at run-time, and fast thread operations, the operating system kernel and a thread management module in a user address space manage threads cooperatively. The authors implemented the cooperative thread management mechanism and measured its performance to examine the effectiveness of the approach.>

A parallel processing scheme and software architecture of SMASH, a parallel processing system for supporting a wide variety of database applications is presented. The main feature of this system is that functional programming concepts are applied to define new database operations and data types and to exploit parallelism inherent in an arbitrary set of database operations. A primitive set (SMASH primitive set) of the software architecture is presented that defines an abstract machine interface between high-level database languages and general-purpose hardware systems for parallel processing. The primitive set is used to implement functional computation systems for executing arbitrary database operations in parallel. A previously proposed stream-oriented parallel processing scheme for relational database operations is extended to support more complex database operations which deal with complex data structures. Several experimental results of parallel processing for database operations are shown to clarify feasibility of the proposed architecture.>

The authors describe a way of noticeably reducing the description cost of database operations executed in distributed computing environments through the design of a declarative language to describe database operations, the development of program transformation techniques to improve efficiency at execution time, and the clarification of prerequisites to execute the programs in distributed computing environments. With the language, database operations are described as functions which manipulate streams. To describe stream manipulation at a higher level, the language SPL (Set Programming Language) is based on mathematical comprehensive notation for sets (ZF expressions). With this language, database operation implementors need not specify any communication primitives; a language processing system automatically translates the programs into procedural programs which include communication primitives.>