Flash SSDs are being incorporated in many enterprise storage platforms recently and expected to play a notable role for IO-intensive applications. However, the IO characteristics of flash SSDs are very different from those of hard disks. Since existent storage subsystems are designed on the basis of characteristics of hard disks, the IO performance of flash SSDs may not be obtained as expected. This paper provides an evaluation of flash SSDs in transaction processing systems with TPC-C benchmark. We present performance results with various configurations and describe our observations of the IO behaviors at different levels along the IO path, which helps to understand the performance of flash-based transaction processing systems and provides certain references to build flash-based systems for IO-intensive applications.

The join operation is one of the most expensive operations in relational database systems. So far many researchers have proposed several hash-based algorithms for the join operation. In a hash-based algorithm, a large relation is first partitioned into several clusters. When clusters overflow, that is, when the size of the cluster exceeds the size of main memory, the performance of hash-based algorithms degrade substantially. Previously we proposed the GN hash algorithm which is robust in the presence of overflown clusters. The GN hash join algorithm combines the Grace hash join and hash-based nested-loop join algorithms. We analyze the performance of the GN hash join algorithm when applied to relations with a non-uniform Zipf-like data distribution. The performance is compared with other hash-based join algorithms: Grace, Hybrid, nested-loop, and simple hash join. The GN hash join algorithm is found to have higher performance on non-uniformly distributed relations. In this paper, the robustness of the GN hash algorithm from the point of choosing a run time method is verified. In the GN hash algorithm, the criterion for selecting a run time method from the two algorithm is determined by using the value calculated from the I/O cost formula of the two algorithms. This criterion cannot be guaranteed to be optimal under every data distribution, that is, the optimal criterion may change depending on the data distribution. When the data distribution is unknown, all data has to be repartitioned in order to get an accurate optimal criterion. However, from the view of choosing a method at run time, it is necessary for the GN hash algorithm to determine an appropriate criterion regardless of the data distribution. Thus, we inspect the criterion adopted in our algorithm under a simulation environment. From simulation results, we find that the range of the criterion is very wide under any data distribution and assure that the criterion determined with the assumption of a uniform data distribution can be used even when the data is highly skewed. Consequently, we can conclude that the GN hash algorithm which dynamically selects the nested-loop and Grace hash algorithms provides good performance in the presence of data skew and its performance is not sensitive to the criterion.

In data centers, large numbers of computers are run simultaneously. These computers consume an enormous amount of energy. Several challenges related to this issue have been published. An energy-efficient storage management method that cooperates with applications was one effective approach. In this method, data and storage devices are managed using application support and the power consumption of storage devices is significantly decreased. However, existing studies do not take the virtualized environment into account. Recently, many data-intensive applications have been run in a virtualized environment, such as the cloud computing environment. In this paper, we focus on a virtualized environment wherein multiple virtual machines run on a physical computer and a data intensive application runs on each virtual machine. We discuss a method for reducing storage device power consumption using application support. First, we propose two storage management methods using application information. One method optimizes the inter-HDD file layout. This method removes frequently-accessed files from a certain HDD and switches the HDD to power-off mode. To balance loads and reduce seek distances, this method separates a heavily accessed file and consolidates files in a virtual machine with low access frequency. The other method optimizes the intra-HDD file layout, in addition to performing inter-HDD optimization. This method places frequently accessed files near each other. Second, we present our experimental results and demonstrate that the proposed methods can create sufficiently long HDD access intervals that power-off mode can be used, and thereby, reduce the power consumption of storage devices.