• The KIPS Transactions:PartD
• /
• v.8D no.5
• /
• pp.593-604
• /
• 2001

A database sharing system (DSS) refers to a system for high performance transaction processing. In the DSS, the processing nodes are coupled via a high speed network and share a common database at the disk level. Each node has a local memory and a separate copy of operating system. To reduce the number of disk accesses, the node caches data pages and index pages in its memory buffer. In general, B-tree index pages are accessed more often and thus cached at more processing nodes, than their corresponding data pages. There are also complicated operations in the B-tree such as Fetch, Fetch Next, Insertion and Deletion. Therefore, an efficient cache consistency scheme supporting high level concurrency is required. In this paper, we propose cache consistency schemes using identifiers of index pages and page_LSN of leaf page. The propose schemes can improve the system throughput by reducing the required message traffic between nodes and index re-traversal.

• The Journal of Information Technology
• /
• v.3 no.3
• /
• pp.117-134
• /
• 2000

Client-server DBMS based on a data-skipping model can exploit client resources effectively by allowing inter-transaction caching. However, inter-transaction caching raises the need of transactional cache consistency maintenance(TCCM) protocol, since each client is able to cache a portion of the database dynamically. Detection-based TCCM schemes can reduce the message overhead required for cache consistency if they validate clients replica asynchronously, and thus they cm show high throughput rates. However, they tend to show high ratios of transaction abort since transactions can access invalid replica. For coping with this drawback, this paper develops a new notion of shadow transaction, which is a backup-purpose one that is kept ready to replace an aborted transaction. This paper proposes a new detection-based TCCM scheme named DL-ST on the basis of the notion of shadow transaction. Using a simulation model, this paper evaluates the effect of shadow transaction in terms of transaction through rate and abort ratio.

• Journal of Digital Convergence
• /
• v.17 no.9
• /
• pp.153-161
• /
• 2019

Discrimination of media object segments in wireless home proxies has a significant impact on caching delay, and caching delay degrades the performance of the proxy. In this paper, we propose a Single Fetching Smart Caching (SFSC) strategy and a Multi-Fetching Smart Caching (MFSC) strategy to improve the proxy performance of the home network and improve the caching performance for media object segments. The SFSC strategy is a technique that performs caching by sequential fetching of object segments requested by the home node one at a time, which guarantees a faster cache hit rate, and the MFSC strategy is a technique that caches the media object segments by blocking object segments requested by the home node one at a time, which improves the throughput of cache. Simulation results show that the cache hit rate and the caching delay are more efficient than the MFSC technique, and the throughput of the object segment is more efficient than that of the SFSC technique.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.4
• /
• pp.15-20
• /
• 2016

Write amplification is a critical factor that limits the stable performance of flash-based storage systems. To reduce write amplification, this paper presents a new technique that cooperatively manages data in flash storage and nonvolatile memory (NVM). Our scheme basically considers NVM as the cache of flash storage, but allows the original data in flash storage to be invalidated if there is a cached copy in NVM, which can temporarily serve as the original data. This scheme eliminates the copy-out operation for a substantial number of cached data, thereby enhancing garbage collection efficiency. Experimental results show that the proposed scheme reduces the copy-out overhead of garbage collection by 51.4% and decreases the standard deviation of response time by 35.4% on average.

• Journal of Digital Convergence
• /
• v.15 no.8
• /
• pp.341-349
• /
• 2017

CCN proposes that intermediate network nodes on a network path for a transmitted data-packet cache the data-packet. If the nodes receive request packets for the cached data, the nodes can directly respond to the request-packets using the cached data. Since a request-packet can be responded by one of the intermediate nodes on a path of the request-packet, both faster response time and decreased data transmission amount are expected comparing to the existing host centric networking. However, CCN is vulnerable against forgery attacks because data-packet receivers cannot identify a data provider. Hence, a data authentication scheme is essentially needed to make CCN more secure. But such a data authentication process is one of the main causes of CCN-based service delays. This paper first analyzes the problems of a CCN data authentication scheme, then proposes an improved authentication operation scheme for efficiently authenticating data, and finally evaluates its performance.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.458-467
• /
• 2001

Optimistic Two-Phase Locking(O2PL) performs as well as or better than the other approaches because it exploits client caching well and also has relatively lower network bandwidth requirements. However, O2PL leads to unnecessary waits, because, it can not be commit a transaction until the transaction obtains all requested locks. In addition, Optimistic Concurrency Control(OCC) tends to make needless aborts. This paper suggests an efficient optimistic cache consistency protocol that overcomes such shortcomings. Our scheme decides whether to commit or abort a transaction without wait and it adopts transaction re-ordering in order to minimize the abort rate. Our scheme needs only one version for each data item in spite of the re-ordering mechanism used. Finally, this paper presents a simulation-based analysis that shows superiority in performance of out scheme to O2PL and OCC.

• Proceedings of the IEEK Conference
• /
• 2003.11c
• /
• pp.223-226
• /
• 2003

In this paper, we propose new agent discovery and route discovery schemes to support Mobile IP (MIP) in Ad Hoc networks with wireless backbone. The wireless backbone consisting of stationary wireless routers and Internet gateways (IGs) is a kind of wireless access network of IP-based core network. The proposed scheme utilizes favorable features of wireless backbone such as stable links and no energy constraints. In the agent discovery scheme, backbone-limited periodic Agent Advertisement (AA) and proxy-AA messages are used, which reduce network-wide broadcasting overhead caused by AA and Agent Solicitation messages and decentralize MIP processing overhead in IGs. In order to reduce delay time and control message overhead during route discovery far the destination outside Ad Hoc network, we propose a cache-based scheme which can be easily added to the conventional on-demand routing protocols. The proposed schemes can reduce control overhead during agent discovery and route discovery, and efficiently support MIP in Ad Hoc network with wireless backbone.

• ETRI Journal
• /
• v.37 no.4
• /
• pp.685-695
• /
• 2015

This paper proposes a novel video delivery scheme that reduces the bandwidth consumption cost from a video server to terminals in Long-Term Evolution networks. This proposed scheme combines optimized hybrid multicast with a segment-based caching strategy for use in environments where the maximum number of multicast channels is limited. The optimized hybrid multicast, allocation of multicast channels, and cache allocation are determined on the basis of a video's request rate, the related video's length, and the variable cost per unit size of a segment belonging to the related video. Performance evaluation results show that the proposed scheme reduces a video's delivery costs. This work is applicable to on-demand TV services that feature asynchronous video content requests.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.8
• /
• pp.719-729
• /
• 2000

Prefetching aims at reducing memory latency by fetching, in advance, data that are likely to be requested by the processor in a near future. The effectiveness of prefetching is determined by how accurate the prediction on the needed instructions and data is. Most previous studies on prefetching were limited to proposing a particular prefetch scheme and its performance evaluation, paying little attention to theoretical aspects of prefetching. This paper focuses on the theoretical aspects of instruction prefetching. For this purpose, we propose a clairvoyant prefetch model that makes use of perfect history information. Based on this theoretical model, we analyzed upper limits on the prefetch prediction accuracies of the SPEC benchmarks. The results show that the prefetch prediction accuracy is very high when there is no cache. However, as the size of the instruction cache increases, the prefetch prediction accuracy drops drastically. For example, in the case of the spice benchmark, the prefetch prediction accuracy drops from 53% to 39% when the cache size increases from 2Kbyte to 16Kbyte (assuming 16byte block size). These results indicate that as the cache size increases, most localities are captured by the cache and that instruction prefetching based on the information extracted from the references that missed in the cache suffers from prediction inaccuracies

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.10
• /
• pp.3262-3271
• /
• 2000

Continuous media objects, due to large volume and real-time consiraints in their delivery,are likely to consume much network andwidth Generally, proxy servers are used to hold the fiequently requested objects so as to reduce the network traffic to the central server but most of them are designed for text and image dae that they do not go well with continuous media data. So, in this paper, we propose a two-layered network cache management policy for continuous media object delivery on the wide area networks. With the proposed cache management scheme,in cach LAN, there exists one LAN cache and each LAN is further devided into a group of sub-LANs, each of which also has its own sub-LAN eache. Further, each object is also partitioned into two parts the front-end and rear-end partition. they can be loaded in the same cache or separately in different network caches according to their access frequencics. By doing so, cache replacement overhead could be educed as compared to the case of the full size daa allocation and replacement , this eventually reduces the backbone network traffic to the origin server.