We propose a new dependable system called PREGMA (Platform for Reliable Environment based on a General-purpose Machine Architecture). PREGMA aims to meet two requirements -- fault tolerance and low cost -- for Internet services. It can provide fault tolerance, so we can avoid system failure and prevent data corruption, even if faults occur. That is, it masks the faults by running multiple replicated servers, each possessing its own data, in a loosely synchronized manner and delivering the majority vote as output to clients. Moreover, PREGMA is composed of COTS (Commercial Off-The-Shelf) components without modification, which makes it possible to offer the services at a low cost. We investigated two approaches for achieving redundancy of the Coordinator, which is the core of PREGMA: using the primary backup method and the active replication method. We evaluated the effectiveness of PREGMA in terms of throughput overhead, data integrity and recovery time. The results for a prototype show that PREGMA using the Coordinator with the primary backup method outperforms that with the active replication method and has throughput only 3% lower than a non-redundant system. The results also show that, in the event of failure, the recovery time is only less than one second and no data corruption occurs.

Personalized PageRank (PPR) is a typical similarity metric between nodes in a graph, and node searches based on PPR are widely used. In many applications, graphs change dynamically, and in such cases, it is desirable to perform ad hoc searches based on PPR. An ad hoc search involves performing searches by varying the search parameters or graphs. However, as the size of a graph increases, the computation cost of performing an ad hoc search can become excessive. In this paper, we propose a method called Castanet that offers fast ad hoc searches of PPR. The proposed method features (1) iterative estimation of the upper and lower bounds of PPR scores, and (2) dynamic pruning of nodes that are not needed to obtain a search result. Experiments confirm that the proposed method does offer faster ad hoc PPR searches than existing methods.

There is a strong demand to expand captioned broadcasting for TV news programs in Japan. However, keyboard entry of captioned manuscripts for news program cannot keep pace with the speed of speech, because in the case of Japanese it takes time to select the correct characters from among homonyms. In order to implement simultaneous subtitled broadcasting for Japanese news programs, a simultaneous subtitling system by speech recognition has been developed. This system consists of a real-time speech recognition system to handle broadcast news transcription and a recognition-error correction system that manually corrects mistakes in the recognition result with short delay time. NHK started simultaneous subtitled broadcasting for the news program "News 7" on the evening of March 27, 2000.

Focusing on a distributed control service-control-node (SCP) that houses a database (DB) distributed across multiple modules, this paper proposes an autonomous distributed SCP architecture using multicasting access to the distributed DB, and highlights its application areas. We assume as a basic condition that neither the network nor the other modules in the system are aware of the DB configuration. Based on this condition, we propose two basic methods: a unicast approach in which the DB management module that is selected at random by the network routes the DB access request to the module where the target data resides (Method A), and a multicast method in which DB access requests are broadcast to all modules (Method B). A quantitative evaluation is made of the number of required modules and required communications performance between modules which is determined by the capacity of the main memory and processing capacity of the processors. Based on the results, we conclude that Method B better exploits the advantages of module autonomous distribution technology within the limits that the economy of inter-module communication overhead is not impaired. Furthermore, in the event a module fails in Method B, a scheme is proposed in which the defective module is cut out of the multicast group, and multicasting continues. This could be implemented most effectively using a separate route under hardware control that is independent of the on-line communications route between modules.

In recent years, dramatic improvements have been made to computer hardware. In particular, the number of cores on a chip has been growing exponentially, enabling an ever-increasing number of processes to be executed in parallel. Having been originally developed for single-core processors, database (DB) management systems (DBMSs) running on multicore processors suffer from cache conflicts as the number of concurrently executing DB processes (DBPs) increases. Therefore, a cache-efficient solution for arranging the execution of concurrent DBPs on multicore platforms would be highly attractive for DBMSs. In this paper, we propose CARIC-DA, middleware for achieving higher performance in DBMSs on multicore processors, by reducing cache misses with a new cache-conscious dispatcher for concurrent queries. CARIC-DA logically range-partitions the dataset into multiple subsets. This enables different processor cores to access different subsets by ensuring that different DBPs are pinned to different cores and by dispatching queries to DBPs according to the data-partitioning information. In this way, CARIC-DA is expected to achieve better performance via a higher cache hit rate for the private cache of each core. It can also balance the loads between cores by changing the range of each subset. Note that CARIC-DA is pure middleware, meaning that it avoids any modification to existing operating systems (OSs) and DBMSs, thereby making it more practical. This is important because the source code for existing DBMSs is large and complex, making it very expensive to modify. We implemented a prototype that uses unmodified existing Linux and PostgreSQL environments, and evaluated the effectiveness of our proposal on three different multicore platforms. The performance evaluation against benchmarks revealed that CARIC-DA achieved improved cache hit rates and higher performance.