Bhum Cheol Lee

Reconfigurable shared buffer ATM switch

This paper describes the architecture of a reconfigurable shared buffer asynchronous transfer mode (ATM) switch and its VLSI implementation. The reconfigurable shared buffer ATM switch on one chip has a shared buffer of 4 ns scalable pipelined memory. It solves the restriction of memory cycle time in a shared buffer ATM switch, and supports flexible switching performance by the scalability of the embedded buffer. The proposed switch provides port size scalability with the independence of queue address control from buffer memory control. The switch size and the buffer size of the proposed ATM switch can be reconfigured without serious circuit redesign. Prototype chip has been designed for 4/spl times/4 ATM switch that has a shared buffer of 128-cell. It is integrated in 10.6/spl times/10.6 mm/sup 2/ with 0.6 /spl mu/m twin well, double-metal, and single-poly CMOS technology. Simulated operating frequency is 80 MHz which supports 640 Mbps per port.

Fault localization in NFV framework

Network function virtualization is quickly gaining acceptance as the new approach to delivering communication services. The promises of greater flexibility and dramatically reduced time to introduce new services coupled with the cost advantages, are driving communications service providers(CSP) around the world to begin deploying NFV-based services. But NFV service is actually operated based on virtualization environment, it has greater impact on disability virtual resources, when the physical devices had a fault. In this paper, when physical faults occurred in the NFV Framework, or logical failure occurs on a specific logical device, we will present a structure for Fault Localization method to ensure the continuity of service in NFVI.

Scalable flow-based network processor for premium network services

This paper presents OmniFlow system that is suitable for future Internet and premium network services requiring multi-layer protocol processing. It is implemented with multi-core processors running a flow-based packet processing for high parallelism performance. It provides for shortening development time, cutting down system cost, and scalability. Its key technology is a flow-based network processor that contains 40 RISC cores, being implemented as ASIC. The network processor is the 40Gbps packet forwarding programmable engine to process incoming packets through two SPI-4.2s.

The architecture of generic VNFM using TOSCA

Network function virtualization is quickly gaining acceptance as the new approach to delivering communication services. The promises of greater flexibility and dramatically reduced time to introduce new services coupled with the cost advantages, are driving communications service providers(CSP) around the world to begin deploying NFV-based services. An NFV operation system is divided into the NFV Orchestrator (NFVo), VNF Manager (VNFM), and VIM (Virtual Infrastructure Manager) subsystems in accordance with the ETSI standard, and the VNFM is defined as the system provided by the VNF manufacturer. In this paper, we present a generic VNF management system for administration and operation of the NVF” under the following structure and to specify the Or-Vnfm, Ve-Vnfm and Vi-Vnfm interfaces, with the goal of developing a generic and scalable VNFM to support various VNFs.

Cloud based Video Storage System with privacy protection

In this paper, we present a novel method to protect privacy information for the Cloud based Video Storage System (CVSS), which not only distribute the CVSS geographically, but also use different privacy protection key algorithm for different CVSS to encode/decode the actual video data itself stored in the storage with the subscriber key for the privacy protection so that the system even under the hacking of the CVSS systems, still provide the privacy protection of the video data. We present how this system stores the video stream data transferred from the network connected camera such as IP CCTV. As the system receives the video stream, it masks the privacy related part, such as facial information or car plate number and so on, of the video stream data with a scrambling key and it also encrypts the scrambled video stream data with an encryption key. In this paper, we also present how the system retrieves the privacy information protected video image requested by either the network control centre or end users. When the network control centre or end user requests the networked video, it retrieves the corresponding video image from the video storage first. Then, it decrypts the image with the decryption key and unscrambles the decrypted image with the unscrambling key. Then it transfers the network video back to the network control center or to the end-user. We present the architecture of the distributed CVSS with the privacy protection for the Cloud based Network Function Virtualization System. We also provide flowchart for receive/transmit operation of the DCVSS. Finally, we present a POC implementation of the distributed CVSS in the Cloud based Network Function Virtualization System.

A distributed cloud based video storage system with privacy protection

In this paper, we present a novel method to protect privacy information for the Cloud based Video Storage System (CVSS), which not only distribute the CVSS geographically, but also use different privacy protection key algorithm for different CVSS to encode/decode the actual video data itself stored in the storage with the subscriber key for the privacy protection so that the system even under the hacking of the CVSS systems, still provide the privacy protection of the video data. We present how this system stores the video stream data transferred from the network connected camera such as IP CCTV. As the system receives the video stream, it masks the privacy related part, such as facial information or car plate number and so on, of the video stream data with a scrambling key and it also encrypts the scrambled video stream data with an encryption key. In this paper, we also present how the system retrieves the privacy information protected video image requested by either the network control centre or end users. When the network control centre or end user requests the networked video, it retrieves the corresponding video image from the video storage first. Then, it decrypts the image with the decryption key and unscrambles the decrypted image with the unscrambling key. Then it transfers the network video back to the network control center or to the end-user. We present the architecture of the distributed CVSS with the privacy protection for the Cloud based Network Function Virtualization System. We also provide flowchart for receive/transmit operation of the DCVSS. Finally, we present a POC implementation of the distributed CVSS in the Cloud based Network Function Virtualization System.

A design of virtual machine aware flow switch in the cloud based network function virtualization system

In this paper, we present a design of Virtual Machine aware flow switch in the Cloud based Network Function Virtualization System, which capable of providing a service at a stable speed even though the number of virtual machines connected to a hypervisor is increased by extracting flow information about each of a plurality of packets and providing network virtualization. In this paper, we also present an architecture of the VM aware flow switch, detailed operation method of it, and the POC of the VM aware flow switch.

Tenant based dynamic processor mapping in the cloud network function virtualization system

In this paper, we present a novel processor mapping method for Cloud Network Function Virtualization System which ensures that network traffic processing of virtual machines belonging to the same tenant is not affected by congestion in network traffic of virtual machines belonging to different tenant. In this paper, we provide a method of dynamically mapping a processor, the method including extracting tenant information on a tenant and information on a virtual machine generated by a Cloud OS or controller; classifying virtual machine queues and processors to process the virtual machine queues by tenant; and dynamically mapping the virtual machine queues onto the processors by tenant. The dynamically mapping of the virtual machine queues onto the processors by tenant may include dynamically mapping the processors to process the VMQs in proportion to either a total number of virtual machine queues belonging to the same tenant, or a total number of virtual machine queues belonging to the same tenant. Finally, we describe the operation of the Tenant based Dynamic Processor Mapping including both spreading process and coalescing process with flow chart.

Analysis the architecture of VNFM (Virtual network function manager)

Network function virtualization is quickly gaining acceptance as the new approach to delivering communication services. The promises of greater flexibility and dramatically reduced time to introduce new services coupled with the cost advantages, are driving communications service providers(CSP) around the world to begin deploying NFV-based services. ETSI ISG has created the industry standards required in a short time to target specific technology areas that industry requires, and many global venders are developing VNFM for managing VNF. In this paper, we examine the standardization trends and areas of development trends, and presents the most basic operation of the initialization procedure for performing the steps of VNFM.