• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.6
• /
• pp.177-181
• /
• 2016

This paper analyzes the host cache effectiveness in full virtualization, particularly associated with journaling of guests. We observe that the journal access of guests degrades cache performance significantly due to the write-once access pattern and the frequent sync operations. To remedy this problem, we design and implement a novel caching policy, called PDC (Pollution Defensive Caching), that detects the journal accesses and prevents them from entering the host cache. The proposed PDC is implemented in QEMU-KVM 2.1 on Linux 4.14 and provides 3-32% performance improvement for various file and I/O benchmarks.

• Journal of KIISE
• /
• v.44 no.3
• /
• pp.232-238
• /
• 2017

Application response time is critical to end-user response time in Android smartphones. Due to the plentiful resources of recent smartphones, storage I/O response time becomes a major key factor in application response time. However, existing storage I/O trace tools for Android and Linux give limited information only for a specific I/O layer which makes it difficult to combine I/O information from different I/O layers, because not helpful for application developer and researchers. In this paper, we propose a novel storage I/O trace tool for Android, called AIOPro (Android I/O profiler). It traces storage I/O from application - Android platform - system call - virtual file system - native file system - page cache - block layer - SCSI layer and device driver. It then combines the storage I/O information from I/O layers by linking them with file information and physical address. Our evaluations of real smartphone usage scenarios and benchmarks show that AIOPro can track storage I/O information from all I/O layers without any data loss under 0.1% system overheads.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.2
• /
• pp.134-145
• /
• 2004

With rapid growth in the amount of data transferred on the Internet, traditional storage systems have reached the limits of their capacity and performance. SAN (Storage Area Network), which connects hosts to disk with the Fibre Channel switches, provides one of the powerful solutions to scale the data storage and servers. In this environment, the maintenance of data consistency among hosts is an important issue because multiple hosts share the files on disks attached to the SAN. To preserve data consistency, each host can execute the disk I/O whenever disk read and write operations are requested. However, frequent disk I/O requests cause the deterioration of the overall performance of a SAN cluster. In this paper, we introduce a SANtopia global buffer manager to improve the performance of a SAN cluster reducing the number of disk I/Os. We describe the design and algorithms of the SANtopia global buffer manager, which provides a buffer cache sharing mechanism among the hosts in the SAN cluster. Micro-benchmark results to measure the performance of block I/O operations show that the global buffer manager achieves speed-up by the factor of 1.8-12.8 compared with the existing method using disk I/O operations. Also, File system micro-benchmark results show that SANtopia file system with the global buffer manager improves performance by the factor of 1.06 in case of directories and 1.14 in case of files compared with the file system without a global buffer manager.

• KIISE Transactions on Computing Practices
• /
• v.22 no.2
• /
• pp.95-100
• /
• 2016

In the current era of big data and cloud computing, the amount of data utilized is increasing, and various systems to process this big data rapidly are being developed. A distributed file system is often used to store the data, and glusterFS is one of popular distributed file systems. As computer technology has advanced, NAND flash SSDs (Solid State Drives), which are high performance storage devices, have become cheaper. For this reason, datacenter operators attempt to use SSDs in their systems. They also try to install glusterFS on SSDs. However, since the glusterFS is designed to use HDDs (Hard Disk Drives), when SSDs are used instead of HDDs, the performance is degraded due to structural problems. The problems include the use of I/O-cache, Read-ahead, and Write-behind Translators. By removing these features that do not fit SSDs which are advantageous for random I/O, we have achieved performance improvements, by up to 255% in the case of 4KB random reads, and by up to 50% in the case of 64KB random reads.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.2
• /
• pp.247-251
• /
• 2010

Using SCM in storage systems introduce new potentials for improving I/O performance and reliability. In this paper, we study the use of SCM as a buffer cache that guarantees transactional unit writes. Our proposed method can improve storage system reliability and performance at the same time and can recover the storage system immediately upon a system crash. The Proposed method is based on the LINUX JBD(Journaling Block Device), thus reliability is equivalent to JBD. In our experiments, the file system that adopts our method shows better I/O performance even while guaranteeing high reliability and shows fast file system recovery time (about 0.2 seconds).

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.93-98
• /
• 2024

With the rapid advance in AI (artificial intelligence) and high-performance computing technologies, deep learning is being used in various fields. Deep learning proceeds training by randomly reading a large amount of data and repeats this process. A large number of files are randomly repeatedly referenced during deep learning, which shows different access characteristics from traditional workloads with temporal locality. In order to cope with the difficulty in caching caused by deep learning, we propose a new sampling method that aims at reducing the randomness of dataset reading and adaptively operating on existing buffer cache algorithms. We show that the proposed policy reduces the miss rate of the buffer cache by 16% on average and up to 33% compared to the existing method, and improves the execution time by up to 24%.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.9-17
• /
• 2009

In this paper, we propose a dynamic load distribution technique at the operating system level in mobile storage systems with a heterogeneous storage pair of a small form-factor and disk and a flash memory, which aims at saving energy consumption as well as enhancing I/O performance. Our proposed technique takes a combinatory approach of file placement and buffer cache management techniques to find how the load can be distributed in an energy and performance-aware way for a heterogeneous mobile storage air of a hard disk and a flash memory. We demonstrate that the proposed technique provides better experimental results with heterogeneous mobile storage devices compared with the existing techniques through extensive simulations.

• Journal of KIISE
• /
• v.41 no.11
• /
• pp.878-884
• /
• 2014

File I/O buffer cache plays an important role in narrowing the wide speed gap between the main memory and the secondary storage. However, data loss or inconsistencies may occur if the system crashes before the data that has been updated in the buffer cache is flushed to storage. Thus, most operating systems adopt a daemon that periodically flushes dirty data to the secondary storage. In this study, we show that periodic flushes account for 30-70% of the total write traffic to storage and remove this inefficiency by implementing a small, non-volatile buffer cache. Specifically, we present space-efficient management techniques, such as delta-write and fragment-grouping, and show that the storage write traffic and throughput can be improved by a margin of 44.2% and 23.6%, respectively, with only a small NVRAM.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.395-401
• /
• 2004

In the current high network-centralized computing and enterprising environment, it is getting essential to transmit data reliably at very high rates. Until now previous client/server model based NFS(Network File System) or AFS(Andrew's Files System) have met the various demands but from now couldn't satisfy those of the today's scalable high-performance computing environment. Not only performance but data sharing service redundancy have risen as a serious problem. In case of NFS, the locking issue and cache cause file system to reboot and make problem when it is used simply as ip-take over for H/A service. In case of AFS, it provides file sharing redundancy but it is not possible until the storage supporting redundancy and equipments are prepared. Lustre is an open source based cluster file system developed to meet both demands. Lustre consists of three types of subsystems : MDS(Meta-Data Server) which offers the meta-data services, OST(Objec Storage Targets) which provide file I/O, and Lustre Clients which interact with OST and MDS. These subsystems with message exchanging and pursuing scalable high-performance file system service. In this paper, we compare the transmission speed of gigabytes file between Lustre and NFS on the basis of concurrent users and also present the high availability of the file system by removing more than one OST in operation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.7
• /
• pp.625-635
• /
• 2008

Portable multimedia players have some different properties compared to general multimedia file server. Some of those properties are single user ownership, relatively low hardware performance, I/O burst by user special functions, and short software development cycles. Though suitable for processing multiple user requests at a time, the general multimedia file systems are not efficient for special user functions such as fast forwards/backwards. Soml' methods has been proposed to improve the performance and functionality, which the application programs give prediction hints to the file system. Unfortunately, they require the modification of all applications and recompilation. In this paper, we present a file system that efficiently supports user special functions in embedded multimedia systems using file block allocation, buffer-cache, and prefetch. A prefetch algorithm, SPRA (SPeed-based PRefetch Algorithm) predicts the next block using I/O patterns instead of hints from applications and it is resident in the file system, so doesn't affect application development process. From the experimental file system implementation and comparison with Linux readahead-based algorithms, the proposed system shows $4.29%{\sim}52.63%$ turnaround time and 1.01 to 3,09 times throughput in average.