• Journal of IKEEE
• /
• v.5 no.1 s.8
• /
• pp.111-120
• /
• 2001

VC merging involves distinguishing cells from an identical merged VC label. Various approaches have been proposed to help this identification process. However, most of them incur additional buffering, protocol overhead and/or variable delay. They make the provision of QoS difficult to achieve. So it was proposed a merge capable scheduler to support VC-merging (VCMS). However, in situations where all VCs are to be merged or the incoming traffic load is very low, it could happen that there are not enough non-merging cells to snoop. In this situation the scheduler uses special control cells to fill the empty time slots out. Too many control cells can cause high cell loss ratio and an additional packet transfer delay. To overcome the drawbacks, we propose a Sequence Skipping(SS) method where the sequencers skip the empty queues and insert SS cells. We show SS method is suitable for VC-merging and can reduce the cell loss ratio and the mean packet transfer delay through simulations.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.7
• /
• pp.33-40
• /
• 2010

The hand over between different radio access networks, especially where their coverage overlaps, suffers various complications since the different access networks provide different service characteristics. In this paper, it is about the system which can control the wireless LAN network management system and the wireless LAN access point of variety vendor which is in different areas. Successful development of mobile sink wireless LAN network management system which supports multi-vendor devices and have center-local interoperability structure. In proposed system, speed of the mobile station give about 10% improvement when the map domain size is 5 than the existing system. Through a data plane and management system of message way, it give a stability system and consistency network.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.4B
• /
• pp.270-281
• /
• 2003

E-PON supports efficiently the increased Internet data traffic and efficient bandwidth assignment function by which the shared upstream bandwidth in the broadband access networks. In this paper, when different types of incoming sources are mapped into QoS-aware source and Best-Effort(BE) source, we propose the Dual-GPS schedulaer in E-PON that has the characteristics to provide delay and delay-constraint application, and maximizes the bandwidth to best-effort service. When transmit upstream data in E-PON, The proposed scheme supports the control capabilities of delay and delay-variation for required QoS flow through the novel bandwidth assignment scheme. Simulation results show our scheme outperforms the conventional bandwidth allocation scheme in controlling the delay and delay-variation of E-PON system.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.479-484
• /
• 2021

This paper proposed and designed a structure to achieve high performance with a small number of cores in GPGPU with SIMT structure. GPGPU for application to mobile devices requires a structure to increase performance compared to power consumption. In order to reduce power consumption, the number of cores decreased, but to improve performance, the size of the warp scheduler for managing threads was set to 4, which was greatly reduced than 32 of general GPGPU. Reducing warp size can reduce the number of idle cycles in pipelines and efficiently apply memory latency to reduce miss penalty when accessing cache memory. The designed GPGPU measured computational performance using a test program that includes floating point operations and measured power consumption through a 28nm CMOS process to obtain 104.5GFlops/Watt as a performance per power. The results of this paper showed about four times better performance per power compared to Tegra K1 of Nvidia

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.370-372
• /
• 2021

According to the Clinical Research Center for Dementia, 840,000 Koreans aged 65 or older had dementia patients, with a prevalence rate of 10.39%. The prevalence rate is one in 10 elderly people, but difficult for families to take care of them all day. Judged that possible to manage the conditions and schedules of elderly people living alone by utilizing AI speaker system where schedule management is stored. This paper implements modules for AI schedulers in patients with dementia. Configured to link AWS, a remote IOT, inside the raspberry pi, and to output the schedule to speakers using a calendar from Google API. Through this study, judged that ease of scheduling will help manage and schedule dementia patients.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.5_6
• /
• pp.307-318
• /
• 2003

In this paper, we design and construct a TMO object group that provides the guaranteed real-time services in the distributed object computing environments, and verify execution power of its model for the correct distributed real-time services. The TMO object group we suggested is based on TINA's object group concept. This model consists of TMO objects having real-time properties and some components that support the object management service and the real-time scheduling service in the TMO object group. Also TMO objects can be duplicated or non-duplicated on distributed systems. Our model can execute the guaranteed distributed real-time service on COTS middlewares without restricting the specially ORB or the of operating system. For achieving goals of our model. we defined the concepts of the TMO object and the structure of the TMO object group. Also we designed and implemented the functions and interactions of components in the object group. The TMO object group includes the Dynamic Binder object and the Scheduler object for supporting the object management service and the real-time scheduling service, respectively The Dynamic Binder object supports the dynamic binding service that selects the appropriate one out of the duplicated TMO objects for the clients'request. And the Scheduler object supports the real-time scheduling service that determines the priority of tasks executed by an arbitrary TMO object for the clients'service requests. And then, in order to verify the executions of our model, we implemented the Dynamic Binder object and the Scheduler object adopting the binding priority algorithm for the dynamic binding service and the EDF algorithm for the real-time scheduling service from extending the existing known algorithms. Finally, from the numerical analyzed results we are shown, we verified whether our TMO object group model could support dynamic binding service for duplicated or non-duplicated TMO objects, also real-time scheduling service for an arbitrary TMO object requested from clients.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.38 no.10
• /
• pp.1-9
• /
• 2001

The MAC (Medium Access Control) protocol is needed for the OLT(Optical Line Termination) to allocate bandwidth to ONUs(Optical Network Units) and ONTs(Optical Network Terminations) dynamically in the ATM PON(Passive Optical Network). With the protocol, the OLT gathers ONUs' informations and provides grants efficiently to each ONU. Two important functions of the MAC protocol is the grant request procedure and the grant distribution algrithm. The latter has the greatest arithmetic portion in the TC(Transmission Convergence) module, occupies a relatively large portion of the overall chip area, has often been the limiting factor in terms of speed, and should be designed to guarantee the quality of service for various traffics. In this paper, we implement the MAC with the FPGA which can allocate grants dynamically according to the queue length information and the number of active ONUs and distribute grants uniformly to minimize the cell delay variation for each ONU. The structure of the MAC scheduler for the dynamic bandwidth assignment has a programmable look-up table. Also, it has a simple structure, the less chip area, and the lower delay time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.12B
• /
• pp.1450-1458
• /
• 2011

The HPcN (Hybrid & high Performance Convergence Network) in research networks means environment which can provide both computing resource such as supercomputer, cluster and network resource to application researchers in the field of medical, bio, aerospace and e-science. The most representative research network in Korea, KREONET has been developing following technologies through the HERO (Hybrid Networking project for research oriented infrastructure) from 200S. First, we have constructed and deployed a control plane technology which can provide a connection oriented network dynamically. Second, the integrated resource management system technology has been developing for reservation and allocation of both computing and network resources, whenever users want to utilize them. In this paper, a testbed network is presented, which is possible to reserve and allocate network resource using the integrated resource management system. We reserve network resource through GNSI (Grid Network Service Interface) messages between GRS (Global Resource Scheduler) and NRM (Network Resource Manager) and allocate network resource through GUNI (Grid User Network Interface) messages between the NRM (network resource manager) and routers, based on reservation information provided from a user on the web portal. It is confirmed that GUNI interface messages are delivered from the NRM to each router at the starting of reservation time and traffic is transmitted through LSP allocated by the NRM.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.3
• /
• pp.73-82
• /
• 2022

Auto-scaling is one of the most important functions for cloud computing technology. Even if the number of users or service requests is explosively increased or decreased, system resources and service instances can be appropriately expanded or reduced to provide services suitable for the situation and it can improves stability and cost-effectiveness. However, since the policy is performed based on a single metric data at the time of monitoring a specific system resource, there is a problem that the service is already affected or the service instance that is actually needed cannot be managed in detail. To solve this problem, in this paper, we propose a method to predict system resource and service response time using a multivariate time series analysis model and establish an auto-scaling policy based on this. To verify this, implement it as a custom scheduler in the Kubernetes environment and compare it with the Kubernetes default auto-scaling method through experiments. The proposed method utilizes predictive data based on the impact between system resources and response time to preemptively execute auto-scaling for expected situations, thereby securing system stability and providing as much as necessary within the scope of not degrading service quality. It shows results that allow you to manage instances in detail.