Quantum computers are evolving at a rapid pace and are considered next-generation computers with high computational capabilities. However, due to the unique characteristics of qubits, state-of-the-art quantum computers are vulnerable to noise caused by qubit instability. To overcome this, highperformance computing (HPC) systems are utilized for quantum circuit simulations to evaluate complex quantum algorithms with great accuracy. However, quantum circuit simulations have high computational demands, and the data volume increases exponentially as the number of qubits increases. In this paper, we propose SWIFTN, a quantum circuit simulation optimization framework for HPC systems with scalability. To achieve this, it enhances parallelism by dividing the tensor networks and distributing them across multiple GPUs and nodes. Additionally, it reduces computational costs by bypassing tasks through intermittent tensor contraction. Finally, to mitigate the degradation in accuracy due to intermittent tensor contraction,SWIFTNperforms amplitude adjustments. We implement and evaluateSWIFTNusing a Perlmutter supercomputer. Our evaluation results using popular quantum algorithm benchmark (i.e., QAOA) shows thatSWIFTNcan improve the performance by$7.85 \times$with 99.997 % accuracy.

Recently, many decentralized IoT systems incorporating IPFS and blockchain have been widely used to share private data and overcome issues in a centralized IoT system such as a single point of failure and loss of data sovereignty. However, the constrained computing power of IoT devices can result in a performance overhead during encryption processes, which involves computationally intensive tasks. In addition, the blockchain involved in the access control mechanism can raise another performance issue and potentially expose sensitive information due to its transparent nature. To address these issues, in this paper, we propose an efficient IoT framework (EnC-IoT) to enhance performance and data protection (i.e., security and privacy) in a decentralized IoT environment. We first introduce an efficient data encryption scheme that shuffles the data order to enhance the security level. Then, it partitions the data into multiple regions and encrypts/decrypts each region in parallel to accelerate the cryptography process. Second, we provide an efficient access control scheme which adopts a tree-based authentication to perform the access control quickly. Moreover, it employs a proxy re-encryption method which allows the decryption of sensitive information without exposing the owner’s private key to improve both security and privacy levels. We implement EnC-IoT with two schemes in IPFS based on a decentralized IoT environment. The experimental results show that EnC-IoT provides a performance improvement of up to 100.7× for uploading and 9.6× for downloading compared with the existing systems.

This paper introduces a novel file system, ScaleDFS, designed to accelerate decentralized file sharing for a private network, leveraging the potential of scaling file management based on directed acyclic graph (DAG) on modern hardware. Specifically, in ScaleDFS, we first design a DAG builder that parallelizes the construction of DAG nodes for a file while preserving critical orders to speed up the uploading process. Second, we introduce a DAG reader that retrieves leaf DAG nodes in parallel without graph traversal assisted by a devised DAG cache to accelerate the downloading process. Finally, we present a DAG remover that rapidly identifies obsolete DAG nodes/data and removes them in parallel to mitigate the garbage collection overhead without compromising consistency. We implement ScaleDFS based on IPFS and demonstrate that ScaleDFS outperforms IPFS by up to 3.7×, 1.8×, and 12.6× in realistic file, private blockchain, and gateway workloads, respectively.

Various scientific applications with different I/O characteristics are executed in HPC systems. However, underlying parallel file systems are unaware of these characteristics of applications, and using a single fixed file layout for all applications can degrade the performance of HPC systems. In this paper, we propose A2FL, an autonomous and adaptive file layout adjustment scheme that optimizes parallel file system configurations by analyzing the access pattern of the applications. The key steps of A2FL are as follows: (1) A2FL initially intercepts the I/O operations of the application, recording their access patterns in real-time. (2) The access patterns are then transformed into a graphical representation used for predicting I/O performance and providing adjustment recommendations. (3) A2FL autonomously adjusts the file layout based on the prediction results, delivering an optimal file layout within the parallel file system. Moreover, we propose A2FL-Compound which analyzes an access pattern by dividing it into smaller components to optimize the file layout in a fine-grained manner. Our evaluations demonstrate that A2FL significantly enhances I/O performance, with improvements of up to 65.9× compared to the default file layout.

Cybersecurity has become an inevitable concern in the healthcare industry due to the rapid growth of the Internet of Health Things (IoHT). The IoHT is revolutionizing healthcare by enabling remote access to hospital equipment, real-time patient monitoring, and urgent alerts to patients and hospitals. However, the convenience of these systems also makes them vulnerable to cyberattacks, with hackers seeking to disrupt health services or extort money through ransomware attacks. Efficiently detecting multiple threats is a challenging task because IoHT generates large temporal data and system log information. In this paper, we propose time series classification models for the identification of potential cyberattacks in IoHT networks. First, we introduce Neighborhood Component Analysis (NCA) with modifications of the regularization parameter to select the vital input features. With the selected features, we propose two LSTM-based models: Directed Acyclic Graph-based Long Short-Term Memory (DAG-LSTM) and Projected Layer-based Long Short-Term Memory (PL-LSTM) for detecting cyberattacks. We evaluate the existing time series classification models (i.e., GRU, LSTM, and Bi-LSTM) and proposed models (i.e., DAG-LSTM and PL-LSTM) using real-world IoHT data. We also validate the models by applying a non-parametric statistical test, Friedman test. Our evaluation results show that the proposed DAG-LSTM achieves the highest accuracy with 99.89% training and 92.04% an average testing accuracy.

As data is becoming more and more important, Internet of Things (IoT) devices are widely used to collect information and process data from various industries, such as finance, autonomous driving, and smart factories. To address the limited computational power of IoT devices in processing real-time data, both edge computing, which utilizes nearby computers with greater computation capabilities, and cloud computing with even more processing power, are widely adopted solutions. As these systems have heterogeneous software and hardware configurations, it can be challenging to understand the behavior of the application from the perspective of different resources. In this article, we propose an efficient logging and monitoring system in large-scale, heterogeneous environments for IoT and edge applications. To do this, our scheme first collects system resource usage data from each compute node using the operating system’s native system analysis tool. Then, it consolidates the system resource usage information from multiple nodes into an integrated database which creates a comprehensive view of the system. Finally, our scheme provides global system resource information in terms of specific jobs and nodes, providing a comprehensive understanding of complex heterogeneous hardware/software stacks. Our evaluation, using IoT and edge workloads in heterogeneous systems, demonstrates the efficiency of logging and monitoring schemes. The average network usage for Windows and Linux is 0.12 and 1.29 kB/s, respectively, resulting in minimal network overhead. In addition, the proposed scheme shows negligible overhead in terms of both runtime (up to 0.73%) and storage (0.0474%).

Distributed file systems (DFSs) are widely used in large scale computing environments such as cloud computing and high performance computing systems (HPC) where thousands of applications are executed simultaneously. To support applications that produce and process a large amount of data, DFSs manage limited storage resources in the system and allocate the resources to the applications. To efficiently utilize limited storage resources, it is important to consider the I/O characteristics of applications as well as the effect of resource sharing among multiple applications in the system. In this paper, we first perform empirical performance analysis and investigate the effect of I/O interference caused by storage resource allocation. Based on our empirical performance analysis, we propose IFLustre towards interference-free and efficient storage allocation in Lustre file system. IFLustre first utilizes previous execution logs to determine the necessary storage resources for the application and predict the throughput and runtime of the application. Then, it allocates storage resources via file system configurations based on the prediction results and the DFS allocation status. This allows IFLustre to allocate sufficient resources to the application while mitigating interference from resource sharing. The experimental results show that IFLustre can improve the performance by up to 47% compared with Lustre.

In HPC systems, the increasing need for a higher level of concurrency has led to packing more cores within a single chip. However, since multiple processes share memory space, the frequent access to resources in critical sections where only atomic operation has to be executed can result in poor performance. In this paper, we focus on reducing lock contention on the memory management system of an HPC manycore architecture. One of the critical sections causing severe lock contention in the I/O path is in the page management system, which uses multiple Least Recently Used (LRU) lists with a single lock instance. To solve this problem, we propose a Finer-LRU scheme, which optimizes the page reclamation process by splitting LRU lists into multiple sub-lists, each having its own lock instance. Our evaluation result shows that the Finer-LRU scheme can improve sequential write throughput by 57.03% and reduce latency by 98.94% compared to the baseline Linux kernel version 5.2.8 in the Intel Knights Landing (KNL) architecture.

As the size of data grows in modern applications, the efficient usage of limited resources is becoming crucial. In order to reorganize large data under memory limitations, many data-intensive applications utilize external sort as a critical component. To streamline external sort, a storage framework is especially needed since the entire dataset must be loaded and flushed a couple of times during the sorting process. Most existing frameworks have attempted to simplify the storage access pattern by associating each thread with a separate storage device. This prevents randomized and concurrent I/O requests, which impose a huge overhead for legacy drives in order to enable the parallelism needed for external sort. However, such regulations excessively restrain the capabilities of NVMe-based SSDs that deliver high throughput with abundant parallelism. In this article, we present a new framework for external sort that exploits both external and internal parallelism. Externally, any number of threads are mobilized to parallel external sort in a scalable way, even with one NVMe SSD. Meanwhile, some arbitration schemes, such as adaptive resource allocation and fairness control, are adopted to preserve the internal efficiency of storage devices. Our evaluation results demonstrate that our scheme can greatly improve both the I/O efficiency and scalability compared to the existing frameworks.

Large-scale high performance computing (HPC) systems typically consist of many thousands of CPUs and storage units, while used by hundreds to thousands of users at the same time. Applications from these large numbers of users have diverse characteristics, such as varying compute, communication, memory, and I/O intensiveness. A good understanding of the performance characteristics of each user application is important for job scheduling and resource provisioning. Among these performance characteristics, the I/O performance is difficult to predict because the I/O system software is complex, the I/O system is shared among all users, and the I/O operations also heavily rely on networking systems. To improve the prediction of the I/O performance on HPC systems, we propose to integrate information from a number of different system logs and develop a regression-based approach that dynamically selects the most relevant features from the most recent log entries, and automatically select the best regression algorithm for the prediction task. Evaluation results show that our proposed scheme can predict the I/O performance with up to 84% prediction accuracy in the case of the I/O-intensive applications using the logs from CORI supercomputer at NERSC.

As flash-based solid-state drive (SSD) becomes more prevalent because of the rapid fall in price and the significant increase in capacity, customers expect better data services than traditional disk-based systems. However, the order of magnitude performance provided and new characteristics of flash require a rethinking of data services. For example, backup and recovery is an important service in a database system since it protects data against unexpected hardware and software failures. To provide backup and recovery, backup/recovery tools or backup/recovery methods by operating systems can be used. However, the tools perform time-consuming jobs, and the methods may negatively affect run-time performance during normal operation even though high-performance SSDs are used. To handle these issues, we propose an SSD-assisted backup/recovery scheme for database systems. Our scheme is to utilize the characteristics (e.g., out-of-place update) of flash-based SSD for backup/recovery operations. To this end, we exploit the resources (e.g., flash translation layer and DRAM cache with supercapacitors) inside SSD, and we call our SSD with new backup/ recovery functionality BR-SSD. We design and implement the functionality in the Samsung enterprise-class SSD (i.e., SM843Tn) for more realistic systems. Furthermore, we exploit and integrate BR-SSDs into database systems (i.e., MySQL) in replication and redundant array of independent disks (RAID) environments, as well as a database system in a single BR-SSD. The experimental result demonstrates that our scheme provides fast backup and recovery but does not negatively affect the run-time performance during normal operation.

In high-performance computing, storage is a shared resource and used by all users with many different application requirements and knowledge of storage. Consequently, the optimal storage configuration varies according to the I/O behavior of each application. While system logs are helpful resources in understanding the storage behavior, it is non-trivial for each user to analyze the logs and adjust complex configurations. Even for experienced users, it is difficult to understand the full stack of I/O systems and find the optimal configuration for the specific application. In this work, we analyzed the I/O activities of CORI which is an HPC system in National Energy Research Scientific Computing Center (NERSC). The result of our analysis shows that most users do not adjust storage configurations and use the default settings. Also, it shows that only a few applications are executed repeatedly in the HPC environment. Based on this result, we have developed DCA-IO, a dynamic distributed file system configuration adjustment algorithm, which utilizes system log information and widely adapted rules to adjust storage configurations automatically without any user intervention. DCA-IO utilizes existing system logs and does not require any modifications in code or an additional library. To demonstrate the effectiveness of DCA-IO, we have performed experiments using I/O kernels of the real applications in both isolated small-sized Lustre environment and CORI. Our experimental result shows that the use of our scheme can lead to improvements in the performance of HPC applications by up to 75% in an isolated environment and 50% in a real HPC environment without user intervention.

Scientific applications frequently spend a large fraction of their execution time in reading and writing data on parallel file systems. Identifying these I/O performance bottlenecks and attributing root causes are critical steps toward devising optimization strategies. Several existing studies analyze I/O logs of a set of benchmarks or applications that were run with controlled behaviors. However, there is still a lack of general approach that systematically identifies I/O performance bottlenecks for applications running "in the wild" on production systems. In this study, we have developed an analysis approach of "zooming in" from platform-wide to application-wide to job-level I/O logs for identifying I/O bottlenecks in arbitrary scientific applications. We analyze the logs collected on a Cray XC40 system in production over a two-month period. This study results in several insights for application developers to use in optimizing I/O behavior.

Backup and recovery is an important feature of database systems since it protects data against unexpected hardware and software failures. Database systems can provide data safety and reliability by creating a backup and restoring the backup from a failure. Database administrators can use backup/recovery tools that are provided with database systems or backup/recovery methods with operating systems. However, the existing tools perform time-consuming jobs and the existing methods may negatively affect run-time performance during normal operation even though high-performance SSDs are used. In this paper, we present an SSD-assisted backup/recovery scheme for database systems. In our scheme, we extend the out-of-place update characteristics of flash-based SSDs for backup/recovery operations. To this end, we exploit the resources (e.g., flash translation layer and DRAM cache with supercapacitors) inside SSDs, and we call our SSD with new backup/recovery features BR-SSD. We design and implement the backup/recovery functionality in the Samsung enterprise-class SSD (i.e., SM843Tn) for more realistic systems. Furthermore, we conduct a case study of BR-SSDs in replicated database systems and modify MySQL with replication to integrate BR-SSDs. The experimental result demonstrates that our scheme provides fast recovery while it does not negatively affect the run-time performance during normal operation.

Generalized processor sharing (GPS) service policy is a scheduling algorithm to allocate the bandwidth of a queueing system with multi-class input traffic. Simulating the GPS system in realistic traffic environments requires a large amount of time. Thus, fluid simulation is useful because it requires much less time. We analyze the bandwidth allocation for fluid simulation in GPS servers, in which the traffic into the server is treated as fluid. Three properties which characterize GPS servers with fluid input are discussed. We show that there exists a unique bandwidth allocation with the properties. It is shown that our previously proposed algorithm gives the unique bandwidth allocation and it is equivalent to the well-known Newton-Rapson method. In numerical study, the performance of finding the unique bandwidth allocation based on other known root finding methods is compared with that of our previous algorithm. We also investigate the impact of the length of the unit time on the accuracy of the performance measures.

SUMMARY: Genomes of more than 60 fungal species have been sequenced to date, yet there has been no systematic approach to analyze fungal transcription factors (TFs) kingdom widely. We developed a standardized pipeline for annotating TFs in fungal genomes. Resulting data have been archived in a new database termed the Fungal Transcription Factor Database (FTFD). In FTFD, 31,832 putative fungal TFs, identified from 62 fungal and 3 Oomycete species, were classified into 61 families and phylogenetically analyzed. The FTFD will serve as a community resource supporting comparative analyses of the distribution and domain structure of TFs within and across species. AVAILABILITY: All data described in this study can be browsed through the FTFD web site at http://ftfd.snu.ac.kr/.

Bandwidth optimization is considered when several classes of Internet traffic are served in generalized processor sharing (GPS) servers. Internet traffic shows self-similar patterns that make it difficult to obtain analytical performance in GPS. Thus, for performance estimation of different classes of traffic, we use fluid simulation techniques that can reduce the simulation complexity, compared to packet-level simulation. Using the relationship between the guaranteed bandwidth vector and the corresponding performance, we propose a bandwidth optimization problem to minimize the total bandwidth such that performance requirements are satisfied. We use an exterior penalty function method to solve the optimization problem. However, a penalized objective function may have local minimum which is not a global minimum. Thus, we propose a new methodology to circumvent the limitation of the exterior penalty function method.

The paper is concerned with estimating the available bandwidth of a network path. We first develop a theory to estimate the available bandwidth of a queueing system. In order to estimate the available bandwidth, we propose a probing method, called a minimal backlogging method, and a statistic based on the service rate of minimally backlogging probing traffic. We show that the available bandwidth of a queueing system can be estimated by the statistic if probing packets are sent to the queueing system by the minimal backlogging method. For a network path consisting of multiple hops, we extend the approach for a single server by introducing a simplified path model. Since the proposed mechanism can estimate the available bandwidth quickly and track it adaptively, a reasonable range of available bandwidth for a short time interval can be obtained using the mean and variance of the estimated available bandwidth. The performance of the proposed available bandwidth estimation mechanism is evaluated by simulation in a multiple hop network topology.