• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.557-560
• /
• 2002

IP층 핸드오프 중에 등록 지연 (latency)과 지연으로 야기된 패킷 손실 (loss)이 존재할 수 있으며, 이 중 지연은 실시간 서비스 또는 지연에 민감한 서비스에 크게 영향을 미친다. 이러한 지연을 해결하기 위해서 IETF는 사전등록 (Pre-Registration)을 할 수 있는 Low Latency Handoff을 제안했다. 여기에서는 이전 외부에이전트 (oFA)와 새로운 외부에이전트 (nFA)가 존재하여야 되며, 만약 nFA가 없다면 사후등록(Post-Registration)을 수행하게 된다. 그러나 다양한 네트워크환경에서 외부에이전트가 존재하지 않은 경우도 있을 수 있다. 본 논문에서는 이러한 상황에 적응하도록 링크계층 정보를 이용하여 사전등록을 할 수 있도록 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.2977-2997
• /
• 2018

In big data era, fresh data grows rapidly every day. More than 30,000 gigabytes of data are created every second and the rate is accelerating. Many organizations rely heavily on real time streaming, while big data stream computing helps them spot opportunities and risks from real time big data. Storm, one of the most common online stream computing platforms, has been used for big data stream computing, with response time ranging from milliseconds to sub-seconds. The performance of Storm plays a crucial role in different application scenarios, however, few studies were conducted to evaluate the performance of Storm. In this paper, we investigate the performance of Storm under different application scenarios. Our experimental results show that throughput and latency of Storm are greatly affected by the number of instances of each vertex in task topology, and the number of available resources in data center. The fault-tolerant mechanism of Storm works well in most big data stream computing environments. As a result, it is suggested that a dynamic topology, an elastic scheduling framework, and a memory based fault-tolerant mechanism are necessary for providing high throughput and low latency services on Storm platform.

• International journal of advanced smart convergence
• /
• v.9 no.3
• /
• pp.253-259
• /
• 2020

Most of today's large-scale cloud systems and enterprise data centers are distributing resources to improve scalability and resource utilization. NVMe-over-Fabric protocol allows submitting NVMe commands to a remote NVMe SSD through RDMA (Remote Direct Memory Access) network. It is attracting attention recently because it is possible to construct a disaggregation storage system with low latency through the protocol. However, the current I/O stack of NVMe-over-Fabric has an inefficient structure for maintaining compatibility with the traditional I/O stack. Therefore, in this paper, we propose a new mechanism to reduce I/O latency and CPU overhead by modifying I/O path of NVMe-over-Fabric to pass through legacy block layer. According to the performance evaluation results, the proposed mechanism is able to reduce the I/O latency and CPU overhead by up to 22% and 24% compared to the existing NVMe-over-Fabrics protocol, respectively.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.7 no.4
• /
• pp.193-199
• /
• 2012

Virtualization allows multiple commodity operating systems to share on a single physical machine. Resource allocation among virtual machines is a key to determine virtual machine performance. To satisfy time-sensitive task on a domain, hypervisor needs to observe the resource requirements and allocates proper amount of CPU resources in a timely manner. In this paper, we propose a realtime credit scheduler for latency sensitive application on virtual machines. The key idea is to register a time event in the Xen hypervisor. Experiment result shows that the proposed scheme is superior to Credit scheduler.

• Sleep Medicine and Psychophysiology
• /
• v.18 no.2
• /
• pp.67-71
• /
• 2011

Schizophrenia is a chronic, currently incurable, and devastating syndrome. Although sleep disturbances are not primary symptoms of schizophrenia, they are important aspects of schizophrenia. Difficulties initiating or maintaining sleep are frequently encountered in patients with schizophrenia. Many schizophrenics report low subjective sleep quality. Measured by polysomnography, increased sleep latency as well as reduced total sleep time, sleep efficiency, slow wave sleep, and rapid eye movement sleep latency (REM latency), are found in most patients with schizophrenia and appear to be an important aspect of the pathophysiology of this disorder. Some literatures suggest that worsening sleep quality precedes schizophrenic exacerbations. Co-morbid sleep disorders such as obstructive sleep apnea (OSA) and restless legs syndrome (RLS), and sleep-disrupting behaviors associated with schizophrenia may lead to sleep disturbances. Clinicians should screen the patient with sleep complaints for primary sleep disorders like OSA and RLS, and carefully evaluate sleep hygiene behaviors of all patients with schizophrenia who complain of sleep disturbances.

• 한국ITS학회:학술대회논문집
• /
• v.2007 no.10
• /
• pp.225-229
• /
• 2007

Recently, wireless sensor networks are deployed in various applications range from simple environment monitoring systems to complex systems, which generate large amount of information, like motion monitoring, military, and telematics systems. Although wireless sensor network nodes are operated with low-power 8bit processor to execute simple tasks like environment monitoring, the nodes in these complex systems have to execute more difficult tasks. Generally, MAC protocols for wireless sensor networks attempt to reduce the energy consumption using duty cycling mechanism which means the nodes periodically sleep and wake. However, in the duty cycling mechanism. a node should wait until the target node wakes and the sleep latency increases as the number of hops increases. This sleep latency can be serious problem in complex and sensitive systems which require high speed data transfer like military, wing of airplane, and telematics. In this paper, we propose a solution for reducing transmission latency through distributed duty cycling (DDC) in wireless sensor networks. The proposed algorithm is evaluated with real-deployment experiments using CC2420DBK and the experiment results show that the DDC algorithm reduces the transmission latency significantly and reduces also the energy consumption.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.6B
• /
• pp.405-413
• /
• 2012

WFQ(Weighted Fair Queuing) is the most popular fair queuing algorithm to guarantee the Quality-of-Service(QoS), but it has the inherent drawback of a poor resource utilization, particularly under the low rate traffic requiring a tight delay bound. It was recently identified that the poor utilization is mainly due to non-optimized latency of a traffic flow and then LOFQ(Latency-Optimized Fair Queuing) to overcome the drawback was introduced. The LOFQ algorithm, however, renews their optimal latencies for all flows whenever a new flow arrives, which results in the high implementation complexity of O($N^2$).This paper is to reduce thecomplexity to O(1). The proposed method is first to derive the optimal latency index function from the statistical QoS characteristics of the offered load, and then to simply calculate the optimal latency index of the arriving flow using the function.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.6C
• /
• pp.601-606
• /
• 2005

Recently with the high expectation of voice over WLAN service, to supped fast inter-AP security transition in WLAN AP is one of the most actively investigating issues. It is also very important to minimize inter-AP security transition latency, while maintaining constantly the secure association from old AP when a station transits to new AP. Hence, this paper first defines secure transition latency as a primary performance metric of AP system in WLAN supporting IEEE802.11i, 802.1x, and 802.11f, and then presents low latency inter-AP security transition mechanism and its security FSM whose objective is to minimize inter-AP transition latency. Experiment shows that the proposed scheme outperforms the legacy 802.1X AP up to $79\%$ with regard to the transition latency.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.11
• /
• pp.581-591
• /
• 2001

We consider some problems of remote file accesses of multiple clients in mobile computing environments. In this environment, user mobility affects file access latency. Mobile hosts have severe resource constraints in terms of limited size of non-volatile storage. Thus, the burden of computation and communication load raise file access latency. In this paper, we propose a scheme for reducing the file access latency through the file migration. The objective is to minimize of file accesses for all mobile hosts which delivering the file to clients as quickly as possible. We develope an on-demand scheme which determines when the file server should migrate files to another server, or when it should transfer files to mobile hosts. Using simulation, we examine the effects which parameters such as file access frequency, file size, mobility rate have on file system access latency. Through simulation results, we show that our proposed migration scheme is effective in reducing the access latency on the requested file of a mobile host with high access rate and low mobility.

• Journal of Microbiology
• /
• v.40 no.1
• /
• pp.1-10
• /
• 2002

At the dose of 1000 p.f.u. per mouse,100% mortality occurred in mice inoculated with wild-type pseudorabies virus (PrV). In contrast, upon stable deletion of 10 bp nucleotides at the Bsrl site within the TK gene, PrV was rendered to be completely apathogenic. The deletion also caused the virus to be less capable of replicating in respiratory as well as in nervous system tissues. Although animals were exposed to high titers of TK-deleted PrVs, the virus failed to replicate to a high titer as compared to the pathogenic parental virus. In contrast to previous studies the deletion in the TK gene did not prevent the virus from establishing latency. Upon immunosuppression, the latent virus? however, reactivated but replicated at low titers. Interestingly, TK-deleted virus established latency and reactivation, that are occurred only in trigeminal ganglia and the cerebrums and no other tissues involved. Following reactivation, there was no indication of virus shedding in respiratory tissues as confirmed by virus isolation and polymerase chain reaction (PCR) technique targeting at the gB gene of PrV, The non-pathogenic virus with non-shedding characteristics, upon reactivation of the latent virus, would be the important feature of a live virus vaccine candidate.