Toshiaki Tarui

A “network-paging” based method for wide-area live-migration of VMs

In cloud-computing environments, migration of virtual machines (VMs) between data centers can solve many problems such as load balancing and power saving. One of the difficulties in wide-area migration, however, is the “address-warping” problem, in which the address of the VM warps from the source server to the destination server. This confuses or complicates the status of the WAN, and the LANs connected to the WAN. We propose two solutions to this problem. One is to switch an address-translation rule, and the other is to switch multiple virtual networks. The former is analogous to paging in memory virtualization, and the latter is analogous to segmentation. The “network-paging” based method is described and our evaluation results are shown. It took less than 100 ms in average to switch from the source to the destination server using this method.

Evaluation of the lock mechanism in a snooping cache

This paper discusses the design concepts of a lock mechanism for a Parallel Inference Machine (the PIM/c prototype) and investigates the performance of the mechanism in detail. Lock operations are extremely frequent on the PIM; however, lock contention rarely occurs during normal memory usage. For this reason, the lock mechanism is designed so as to minimize the lock overhead time in the case of no contention. This is done by using an invalidation lock mechanism, which utilizes the exclusive state of the snooping cache and in which the locked address is not broadcast. Experimental results demonstrate the benefits of the lock mechanism in regions of few lock contentions. They also confirm that, in most cases, the lock mechanism works well on the PIM. However, the mechanism is also found to cause performance degradation when a locked address is accessed by multiple processing elements (PEs) in a tightly-coupled multi-processor (TCMP). This is because shared data such as the flags for inter-PE communication, which are shared by all the PEs, may be accessed by multiple PEs at the same time, thus generating heavy contention. This paper also shows that combining a register-based broadcasting facility with the proposed lock mechanism can solve the above problem.

Federating heterogeneous network virtualization platforms by slice exchange point

An architecture called the slice-exchange-point (SEP) has been designed for federating heterogeneous net-work-virtualization platforms by creating and managing slices (virtual networks). SEP enables whole inter-domain resources to be managed by the network manager of any single domain. Slice-operation commands are propagated to other domains through SEP by using a common API. SEP introduces the following four features: infrastructure neutrality, single interface federation, abstract and clean federation, and extensibility of capabilities. SEPs functions to achieve these features are discussed. SEP was partially implemented on two VNode domains and one ProtoGENI domain and was verified to function effectively.

Investigating suitability for server virtualization using business application benchmarks

Server virtualization is now required for data center systems to reduce the number of servers. However, it is still unclear which business applications are suitable for virtualization. We present our evaluation results for four types of business application benchmarks on our virtualization system. The results show that the virtualization performance of a TPC-H workload, which mainly executes referencing on a database, uses about 90% of the non-virtualized performance, and that the virtualization performance of the TPC-H workload is better than that of the other benchmark applications. The results of a new performance characteristic for virtualization indicated that application programs, which have performance bottlenecks in disk I/Os and low CPU utilizations in a non-virtualized environment, are suitable for virtualization.

Federation-less federation of ProtoGENI and VNode platforms

Our previous work enabled “federation-less federation”, which means a federation of multiple network-virtualization platforms that do not support federation functions, and applied this method to a homogeneous federation of platforms called the “VNode” infrastructures. In this study, this method was applied to a heterogeneous federation of the ProtoGENI and the “VNode”. We intended to federate these platforms through a single management interface. However, the federation architecture of GENI, which is called the slice-based federation architecture (SFA), cannot be used for single-interface federation but we could not modify the ProtoGENI platform to enable it. Therefore, a method for applying federation-less-federation to ProtoGENI was developed. It enabled federation of these platforms by adding several nodes but without modifying preexisting platforms. This method was applied to federation of the ProtoGENI platform at the University of Utah and two VNode infrastructures in Japan, the slice creation and deletion time was measured and evaluated to be acceptable. Although this federation-less-federation implementation still has several minor problems, it was proved to be useful for experiments and demonstrations.

Evaluation of the Cluster Structure on the PIM/C Parallel Inference Machine

The characteristics of a cluster-structure parallel computer are analyzed and evaluated on the PIM/c parallel inference machine, which consists of eight-processor shared-memory clusters communicating through a processor connected to a network. To avoid communication bottlenecks, the maximum number of processors in a cluster is limited by the ratio of communication operations to program-execution operations. Since this ratio can be as high as 30% on the PIM/c, the network receiving operations should be distributed to processors in the same cluster.

Proposal of virtual network configuration acquisition function for data center operations and management system

Virtualization technologies have been widely deployed in data centers to improve the system utilization. However, they cause increased workload for operators to clarify the structure of virtual networks in data centers. To reduce the operation time, this paper provides the virtual-network management system which automates the integration of the configurations of the virtual networks. The proposed system collects the configurations from server virtualization platforms and VLAN-supported switches, and integrates these configurations according to the newly developed XML-based management information model for virtual-network configurations. The preliminary evaluations show that the proposed system helps to reduce the time to collect and update the configurations by about 40 percent. This result implies that the proposed system is effective for improving the configuration management process for virtual networks in data centers.

VPDC: virtual private data center: a flexible and rapid workload-management system

Rapid server allocation implemented on a virtual private data center (VPDC), which is an autonomous server allocation system for a three-tier Web system, has been developed and tested. The test results show that, with this new system, elapsed time for application server allocation is about 20 seconds, and that for database server allocation is within 140 seconds.

VPDC: Virtual Private Data Center

Rapid server allocation implemented on Virtual Private Data Center (VPDC), which is an autonomous server allocation system for a three-tier web system, has been developed and tested. The test results show that with this new system elapsed time for application server allocation is about 20 seconds, and that for database server allocation is 140 seconds.