Masahiro Kawato

Redundant virtual machine placement for fault-tolerant consolidated server clusters

Consolidated server systems using server virtualization involves serious risks of host server failures that induce unexpected downs of all hosted virtual machines and applications. To protect applications requiring high-availability from unpredictable host server failures, redundant configuration using virtual machines can be an effective countermeasure. This paper presents a virtual machine placement method for establishing a redundant configuration against host server failures with less host servers. The proposed method estimates the requisite minimum number of virtual machines according to the performance requirements of application services and decides an optimum virtual machine placement so that minimum configurations survive at any k host server failures. The evaluation results clarify that the proposed method achieves requested fault-tolerance level with less number of hosting servers compared to the conventional N+M redundant configuration approach.

Just-in-Time Server Provisioning Using Virtual Machine Standby and Request Prediction

Server provisioning is a practical technique to reconfigure a shared server and to improve resource utilization of servers in datacenters and enterprise systems. For the complex systems, however, long process of server provisioning impedes prompt solutions to system problems. This paper proposes a technique to shorten the provisioning processing time after the occurrence of the provisioning request by speculative provisioning execution on a virtual machine as standby. In order to start the provisioning execution in advance, a prediction method for the provisioning request is required. This paper presents a prediction model based on the logistic regression model using system performance metrics. From the evaluation using the actual performance data of enterprise systems, for 50% of the server provisioning requests, the provisioning processing time after the request is shorten over 10 minutes by using the 20-minutes look-ahead request prediction model.

Automatic synthesis of SRN models from system operation templates for availability analysis

In order to cost-effectively verify whether system designs of information systems satisfy availability requirements, it is reasonable to utilize a model-based availability assessment of system design containing administrative operation procedures and a system configuration, because it does not require installing and testing in a real environment. However, since the model-based availability assessments typically require special expertise in mathematical modeling, it would be difficult for a practical system designer to build a correct availability model to assess his/her system design. Although there have been several methods to automatically synthesize the availability model from widely-used design description languages, the synthesized models do not capture impacts caused by operations in operation procedures on availability. To address this issue, this paper proposes a method to automatically synthesize an availability model in the form of stochastic reward net (SRN) from Systems Modeling Language (SysML) diagrams to specify operation procedures and system configurations. Modeling all the features of individual operations is impractical because the amount of required information in SysML diagrams input by system designers becomes larger as the number of features increases. To design the availability models with a smallest possible number of features, we classify typical availability-related features of operations into operation templates. The feasibility of the proposed method is studied by a case study based on a real system of a local government. We succeeded in synthesizing the availability models from the SysML diagrams based on an operation procedure and system configuration of the real system, and analyzing the synthesized availability models with an existing model analysis tool.

DIGITAL WATERMARKING OF VIRTUAL MACHINE IMAGES

The widespread use of server and desktop virtualization technologies increases the likelihood of unauthorized and uncontrolled distribution of virtual machine (VM) images that contain proprietary software. This paper attempts to address this issue using a platform-independent digital watermarking scheme applicable to a variety of VM images. The scheme embeds a watermark in the form of files in a VM image; the watermarked VM image is identified based on the embedded files. To reduce the possibility of discovery by an attacker, the names of the embedded files are very similar to the names of pre-existing files in the VM image. Experiments indicate that the approach is fast and accurate, with average turnaround times of 24.001 seconds and 7.549 seconds for watermark generation and detection, respectively.

Guarantee of Freshness in Resource Information Cache on WSPE: Web Service Polling Engine

Managing resource information using web services is an important issue in grid computing. Although some standards for grid computing based on web services have been published, the overhead of SOAP/HTTP and XML processing may cause performance problems in the resource information service for the grid. This paper proposes a scheduling update mechanism for resource information cache to meet the performance requirements of the resource information service. We have designed and implemented this mechanism on an information service named Web Service Polling Engine (WSPE) that collects resource information in accordance with WSDM/MUWS standards. From the results of evaluations, we conclude that WSPE can guarantee the freshness of resource information with a stable load on the WSPE server.

CIM-Based Resource Information Management for Integrated Access Control Manager

An architecture for CIM-based integrated access control middleware is proposed. The proposed architecture employs CIM standards for managing several access control modules for different resources uniformly in consolidated server environments. CIM standards allow the user of the middleware to look up the target resource information through the common interface, to describe abstract policy with grouping the same type of resources together, and to translate the abstract policy to detailed configurations for each access control module automatically. We evaluated the feasibility of the proposed architecture by a pilot implementation for file access control systems. According to the findings of the evaluation, we propose an extension of the CIM_Directory class to improve operations for exploring directories on user interfaces of the middleware.

Resource Information Cache Update Control for Scalable Access Control Management Systems

In private clouds that host many enterprise applications, scalable security management has become an important issue. In our previous work, we had developed integrated access control manager that manages access permissions to a large number of various resources using resource information provided by a resource information service. To improve performance of the resource information service, we introduced a resource information cache and a proactive cache update control method. To avoid overload of the management server due to updating cached information, the proposed method selects a part of cached information by content priority as an update target. In this work, we evaluated the query response time of the resource information service in a third-party enterprise system using the search queries issued during system operations by an administrator. The proposed method reduced average query response time by 35% compared to a conventional reactive update control method.

Renovating legacy distributed systems using virtual appliance with dependency graph

Legacy distributed systems hosted on old infrastructures can be renovated using virtual appliance that is a package of virtual machines, installed applications and their configurations. By converting a legacy distributed system to a virtual appliance, we can conserve the entire system and restart the application on the specific virtualization platforms. However, in order to execute the application service properly on the new hosting environment, some additional network configurations are required to resolve the dependencies inherited in the original system. In this paper, we propose a framework named DS Renovator that converts a legacy distributed system to a virtual appliance and renovates the system on a new hosting environment with optimum deployment for resolving the dependencies. In the virtual appliance generation process, DS Renovator analyzes server dependencies inherent in the legacy distributed system and generates a dependency graph that consists of nodes and edges representing servers and their dependencies respectively. In the virtual appliance deployment process, DS Renovator applies graph partitioning algorithm to the dependency graph to determine the optimum virtual machine placement which minimizes the dependencies between the hosting servers under the capacity limitations. As a graph partitioning algorithm, we propose an edge-contraction based efficient algorithm. The performance of the proposed algorithm is evaluated with case studies in comparison to other approximation algorithms.

Automatic Cache Update Control for Scalable Resource Information Service with WS-Management

Scalable and interoperable security management has become an important issue in advanced datacenter services. In our previous work, we had developed integrated access control manager (IAM) that manages accesses to various resources. To integrate access controls for various resources, a scalable and interoperable resource information service that provides target resource information to compose access control policies is required. To improve interoperability and performance of the resource information service, we brought a resource information cache into the IAM and introduced an automatic cache update control method that updates cached information proactively based on WS-Management and CIM standards. To reduce server loads to update cached information, the proposed method selects a part of cached information by content priority as an update target instead of updating all cached contents. We evaluated the average query response time of resource information service by simulating the proposed method. The proposed method reduced average query response time by 66.66% compared to the conventional reactive update control method.

A sufficiency-based resource reallocation mechanism for Internet-scale computer resource sharing

In systems from which multiple resource providers dynamically provide computer resources over the Internet, a decentralized resource allocation mechanism is essential. In such an environment, resource providers and resource users have different requirements. Satisfying the requirements of both is difficult because there is no centralized resource broker. To resolve this issue, we propose a sufficiency-based, on-demand resource reallocation mechanism that permits reallocation of resources when a current resource user has a higher utility level than that of a resource requestor. We have implemented this mechanism and used it to allocate resources in a simulated environment. Our experimental results have shown that sufficient resources were provided to 48 percent more users by applying our mechanism as compared with a similar mechanism without reallocation