• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.69-76
• /
• 2014

In this paper, we propose a buffering scheme to improve the performance in P2P live streaming systems by adjusting the ratio of caching and prefetching portion of each peer. To this end, we assign all the peers into many groups depending on their playback periods. We then determine the ratio of caching and prefetching portion in each peer depending on its playback time position relative to those of other peers within the same group. In other words, as the playback position of a peer gets later, we increase the ratio of its caching portion. On the contrary, as the playback position of a peer gets eariler, we increase the ratio of its prefetching portion. This can significantly increase the degree of data duplication among peers that belong to each specific group. By simulation experiments, we show that our proposed an integrated prefetching/caching scheme can improve the performance considerably in terms of jitter ratio, initial playback delay and shared buffermap ratio when compared to the existing fixed portion buffering scheme.

• 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

• Journal of Korea Society of Industrial Information Systems
• /
• v.4 no.4
• /
• pp.32-39
• /
• 1999

Video-On-Demand servers have to provide timely processing guarantees and reduce the storage and reduce the storage and bandwidth requirements for continuous media However, compression techniques used in Video-On-Demand servers make the bit rates of compressed video data significantly variable from frame to frame Consequently, most pervious Video-On-Demand servers which use constant bit rate retrieval to guarantee deterministic service under-utilize the system resources and restrict the number of clients. In this paper, I propose a buffer management scheme called CAP(Caching And Prefetching) for Video-On-Demand server using variable bit rate continuous media. By caching and prefetching the data CAP reduces the variation of the compressed data and increases the number of clients simultaneously served and maximizes the utilization of system resources. Results of trace-driven simulations show the effectiveness of the scheme.

• ETRI Journal
• /
• v.25 no.5
• /
• pp.275-287
• /
• 2003

We present a high performance cache structure with a hardware prefetching mechanism that enhances exploitation of spatial and temporal locality. Temporal locality is exploited by selectively moving small blocks into the direct-mapped cache after monitoring their activity in the spatial buffer. Spatial locality is enhanced by intelligently prefetching a neighboring block when a spatial buffer hit occurs. We show that the prefetch operation is highly accurate: over 90% of all prefetches generated are for blocks that are subsequently accessed. Our results show that the system enables the cache size to be reduced by a factor of four to eight relative to a conventional direct-mapped cache while maintaining similar performance.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.2
• /
• pp.1-9
• /
• 2010

The increasing popularity of mobile devices, such as cellular phones, smart phones, and PDAs, has fostered the need to recommend more effective information in ubiquitous environments. We propose the recommendation method for mobile contents service using contexts and prefetching in ubiquitous environment. The proposed method enables to find some relevant information to specific user's contexts and computing system contexts. The prefetching has been applied to recommend to user more effectively. Our proposed method makes more effective information recommendation. The proposed method is conceptually comprised of three main tasks. The first task is to build a prefetching zone based on user's current contexts. The second task is to extract candidate information for each user's contexts. The final task is prefetch the information considering mobile device's resource. We describe a new recommendation.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.9
• /
• pp.461-468
• /
• 2001

Though shared virtual memory (SVM) system promise low cost solutions for high performance computing they suffer from long memory latencies. These latencies are usually caused by repetitive invalidations on shared data. Since shared data are accessed through synchronization and the patterns by which threads synchronizes are repetitive, a prefetching scheme bases on such repetitiveness would reduce memory latencies. Based on this observation, we propose a prefetching technique which predicts future access behavior by analyzing access history per synchronization variable. Our technique was evaluated on an 8-node SVM system using the SPLASH-2 benchmark. The results show the our technique could achieve 34%~45% reduction in memory access latencies.

• Journal of KIISE:Databases
• /
• v.34 no.5
• /
• pp.448-464
• /
• 2007

Location-Based Services (LBSs) provide results to queries according to the location of the client issuing the query. In LBS, techniques such as caching and prefetching are effective approaches to reducing the data transmission from a server and query response time. However, they can lead to cache inefficiency and network overload due to the client's mobility and query pattern. To solve these drawbacks, we propose a semantic prefetching (SP) scheme using prefetching segment concept and improved cache replacement policies. When a mobile client enters a new service area, called semantic prefetching area, proposed scheme fetches the necessary semantic information from the server in advance. The mobile client maintains the information in its own cache for query processing of location-dependent data (LDD) in mobile computing environment. The performance of the proposed scheme is investigated in relation to various environmental variables, such as the mobility and query pattern of user, the distributions of LDDs and applied cache replacement strategies. Simulation results show that the proposed scheme is more efficient than the well-known existing scheme for range query and nearest neighbor query. In addition, applying the two queries dynamically to query processing improves the performance of the proposed scheme.

• Spatial Information Research
• /
• v.21 no.4
• /
• pp.83-89
• /
• 2013

Recently various contents have been produced using the techniques that require high-performance computing process. A lot of services have been being producted as AR(Augmented Reality) service being combined with mobile information service that a moving user search various information based on one's location with. Mobile information service has a characteristic that it needs to get new information according to the location an user moves to. The characteristic requires a lot of communications when user search information moving to a different location. In order to make up for this drawback, we propose a prefetching technique based on speed and viewing angle in this paper. Existing prefetching techniques retrieve the following location of users considering moving speed and direction of the users. The data showed on the screen in AR is limited by the viewing angle of the mobile device. Due to the problems we discussed above, existing prefetching techniques have a demerit that they retrieve a lot more data than needed actually. We propose more efficient way of retrieving data with AR using the viewing angle of the mobile device. The method we propose reduces retrieval of unnecessary location using the users' speed, direction and viewing angle. This method is more efficient than the existing ways of retrieval because we don't need as many data.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.81-93
• /
• 2023

The popularity of short-form video platforms like TikTok has increased recently. Short-form videos are significantly shorter than traditional videos, and viewers regularly switch between different types of content to watch. Therefore, a successful prefetching strategy is essential for this novel type of video. This study provides a resource-effective prefetching technique for streaming short-form videos. The suggested solution dynamically adjusts the quantity of prefetched video data based on user viewing habits and network traffic conditions. The results of the experiments demonstrate that, in comparison to baseline approaches, our method may reduce data waste by 21% to 83%, start-up latency by 50% to 99%, and the total time of Re-buffering by 90% to 99%.

• Annual Conference of KIPS
• /
• 2020.11a
• /
• pp.12-15
• /
• 2020

Data access characteristics can directly affect the efficiency of the system execution. This research is to design an accurate predictor by using historical memory access information, where highly accessible data can be migrated from low-speed storage (SSD/HHD) to high-speed memory (Memory/CPU Cache) in advance, thereby reducing data access latency and further improving overall performance. For this goal, we design a locally weighted linear regression prefetch scheme to cope with irregular access patterns in large graph processing applications for a DARM-PCM hybrid memory structure. By analyzing the testing result, the appropriate structural parameters can be selected, which greatly improves the cache prefetching performance, resulting in overall performance improvement.