Digital Library[ Search Result ]
Distributed Storage System for Reducing Write Amplification on Non-Volatile Memory
http://doi.org/10.5626/JOK.2020.47.2.129
Recently, research on non-volatile memory, such as 3DXpoint, in distributed storage systems has received considerable interest from both academia and industry. However, in order to utilize these state-of-the-art non-volatile memory devices effectively in distributed storage systems, there is a need for improvements in traditional architectures of HDD/SSD-based storage systems. This is because current distributed storage system structures use a dedicated space for journaling to make up for slow storage performance. Also, considering the performance characteristics of non-volatile memory, which are similar to that of DRAM, current distributed storage system structures are not only inefficient in terms of overall performance but also cause write amplification. In this paper, we propose an architecture that mitigates the effects of write amplification in non-volatile memory-based distributed storage systems. To evaluate the proposed architecture and scheme, we have conducted diverse experiments in a CEPH storage system environment. Through these experiments, we have confirmed that the DAXNJ structure proposed in this paper decreases write amplification by 61% during 1M object write operations and increases the overall system performance by 15%.
ESS Operation Scheduling Scheme Using LSTM for Peak Demand Reduction
Yeongung Seo, Seungyoung Park, Myungjin Kim, Sungbin Lim
http://doi.org/10.5626/JOK.2019.46.11.1165
In recent years, blackouts have become more likely in South Korea as the peak demand has sharply increased. In order to address this issue, an energy storage system (ESS) operation scheduling technique has been investigated for its ability to reduce the peak demand by utilizing the power stored in the ESS. If the power demand information is known in advance, an optimal ESS operation scheduling technique can be applied in consideration of both the power stored in the ESS and the power demand to be generated in the future. However, it is difficult to predict the peak demand in advance because it only occurs in a relatively short time period, and the instance of its occurrence differs substantially from day-to-day. Therefore, it is very difficult to implement an optimal ESS operation scheduling technique that requires exact information on power demands in advance. Thus, in this paper, we proposed an ESS operation scheduling method with which to reduce the peak demand by using only historical power demands. Specifically, we employed a long short-term memory (LSTM) network and trained it using the historical power demands and their corresponding optimal ESS discharge powers. Then, we applied the trained network to approximate the optimal ESS operation scheduling. We showed the validity of the proposed method through computer simulations using historical power demand data from four customers. In particular, it was shown that the proposed scheme reduced the peak demand per year by up to about 82.42% compared to the optimal scheme that is only feasible when the exact future power demands are available.
Improving Performance of Flash Storage Using Restricted Copyback
Duwon Hong, Seulgi Shin, Jihong Kim
http://doi.org/10.5626/JOK.2019.46.8.726
In case of modern flash-based SSDs, the performance overhead of internal data migrations is dominated by the data transfer time and not by the flash program time as in old SSDs. In order to mitigate the performance impact of data migrations, we propose rcopyback, a restricted version of copyback. Rcopyback works in a manner similar to the original copyback except that only n consecutive copybacks are allowed. By limiting the number of successive copybacks, the version guarantees internal migration of data using rcopyback without any reliability problem. In order to take a full advantage of rcopyback, we developed a rcopyback-aware FTL, rcFTL, which intelligently decides whether rcopyback should be used or not by exploiting varying host workloads. Our evaluation results show that rcFTL can improve the overall I/O throughput by 54% on average over an existing FTL which does not use copybacks.
I/O Completion Technique of Virtualized System Considering CPU Usage with High-Performance Storage Devices
Hyeji Lee, Taehyung Lee, Minho Lee, Yongju Song, Young Ik Eom
http://doi.org/10.5626/JOK.2019.46.7.612
Recently, the advent of high-performance storage devices such as Samsung Z-SSD and Intel Optane SSD has shifted the I/O systems’ performance overhead from the storage devices to the software I/O layer. To optimize the I/O performance of high-performance storage devices, the hypervisor and operating system have focused on the effectiveness of polling technique, which is one of the I/O completion techniques applied in virtualized systems, and new techniques such as hybrid and adaptive polling are being adopted. This paper reveals the problem of the existing adaptive polling techniques provided by QEMU-KVM hypervisor and proposes a new I/O completion technique, which saves on CPU usage while fully utilizing high-performance storage devices. Our evaluation indicates that the proposed technique reduces CPU usage by up to 39.7% while delaying I/O latency to less than 5.3% only, in comparison to conventional systems.
An Efficient SLC-buffer Management Scheme for TLC NAND Flash-based Storage
Kirock Kwon, Dong Hyun Kang, Young Ik Eom
http://doi.org/10.5626/JOK.2018.45.7.611
In recent years, almost all consumer devices have adopted NAND flash storage as their main storage, and their performance and capacity requirements are getting higher. To meet these requirements, many researchers have focused on combined SLC-TLC storage consisting of high-speed SLC and high-density TLC. In this paper, we redesign the internal structure of the combined SLC-TLC storage to efficiently manage the SLC region inside the storage and propose a scheme that improves the performance of the storage by employing the I/O characteristics of file system journaling. We implemented our scheme on the real storage platform, the OpenSSD jasmine board, and compared it with the conventional techniques. Our evaluation results show that our technique improves the storage performance by up to 65%, compared with the conventional techniques.
A Cross Layer Optimization Technique for Improving Performance of MLC NAND Flash-Based Storages
Jisung Park, Sungjin Lee, Jihong Kim
http://doi.org/10.5626/JOK.2017.44.11.1130
The multi-leveling technique that stores multiple bits in a single memory cell has significantly improved the density of NAND flash memory along with shrinking processes. However, because of the side effects of the multi-leveling technique, the average write performance of MLC NAND flash memory is degraded more than twice that of SLC NAND flash memory. In this paper, we introduce existing cross-layer optimization techniques proposed to improve the performance of MLC NAND flash-based storages, and propose a new integration technique that overcomes the limitations of existing techniques by exploiting their complementarity. By fully exploiting the performance asymmetry in MLC NAND flash devices at the flash translation layer, the proposed technique can handle many write requests with the performance of SLC NAND flash devices, thus significantly improving the performance of NAND flash-based storages. Experimental results show that the proposed technique improves performance 39% on average over individual techniques.
Multi-core Scalable Real-time Flash Storage Simulation
Hyeon-gyu Lee, Sang Lyul Min, Kanghee Kim
http://doi.org/10.5626/JOK.2017.44.6.566
As NAND flash storage is being widely used, its simulation methodologies have been studied in various aspects such as performance, reliability, and endurance. As a result, there have been advances in NAND flash storage simulation for both functional modeling and timing modeling. However, in addition to these advances, there is a need to drastically reduce the long simulation time that is required to evaluate the aging effect on flash storage. This paper proposes a so-called multi-core scalable real-time flash storage simulation method, which can control the simulation speed according to the user’s preference. According to this method, it is possible to speed up the simulation in proportion to the number of CPU cores arbitrarily given while guaranteeing the correctness of the simulation result. Using our simulator implemented in the form of the Linux kernel module, we demonstrate the multi-core scalability and correctness of the proposed method.
An Analysis of the Overhead of Multiple Buffer Pool Scheme on InnoDB-based Database Management Systems
Yongju Song, Minho Lee, Young Ik Eom
The advent of large-scale web services has resulted in gradual increase in the amount of data used in those services. These big data are managed efficiently by DBMS such as MySQL and MariaDB, which use InnoDB engine as their storage engine, since InnoDB guarantees ACID and is suitable for handling large-scale data. To improve I/O performance, InnoDB caches data and index of its database through a buffer pool. It also supports multiple buffer pools to mitigate lock contentions. However, the multiple buffer pool scheme leads to the additional data consistency overhead. In this paper, we analyze the overhead of the multiple buffer pool scheme. In our experimental results, although multiple buffer pool scheme mitigates the lock contention by up to 46.3%, throughput of DMBS is significantly degraded by up to 50.6% due to increased disk I/O and fsync calls.
A Secure and Practical Encrypted Data De-duplication with Proof of Ownership in Cloud Storage
Cheolhee Park, Dowon Hong, Changho Seo
In cloud storage environment, deduplication enables efficient use of the storage. Also, in order to save network bandwidth, cloud storage service provider has introduced client-side deduplication. Cloud storage service users want to upload encrypted data to ensure confidentiality. However, common encryption method cannot be combined with deduplication, because each user uses a different private key. Also, client-side deduplication can be vulnerable to security threats because file tag replaces the entire file. Recently, proof of ownership schemes have suggested to remedy the vulnerabilities of client-side deduplication. Nevertheless, client-side deduplication over encrypted data still causes problems in efficiency and security. In this paper, we propose a secure and practical client-side encrypted data deduplication scheme that has resilience to brute force attack and performs proof of ownership over encrypted data.
In-Memory File System Backed by Cloud Storage Services as Permanent Storages
Kyungjun Lee, Jiwon Kim, Sungtae Ryu, Hwansoo Han
As network technology advances, a larger number of devices are connected through the Internet. Recently, cloud storage services are gaining popularity, as they are convenient to access anytime and anywhere. Among cloud storage services, object storage is the representative one due to their characteristics of low cost, high availability, and high durability. One limitation of object storage services is that they can access data on the cloud only through the HTTP-based RESTful APIs. In our work, we resolve this limitation with the in-memory file system which provides a POSIX interface to the file system users and communicates with cloud object storages with RESTful APIs. In particular, our flush mechanism is compatible with existing file systems, as it is based on the swap mechanism of the Linux kernel. Our in-memory file system backed by cloud storage reduces the performance overheads and shows a better performance than S3QL by 57% in write operations. It also shows a comparable performance to tmpfs in read operations.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr