Dong-Jae Kang

Cloud service broker portal: Main entry point for multi-cloud service providers and consumers

Cloud-service-broker needs a virtual service portal between multiple cloud-service-providers and cloud-service-consumers. The cloud-service-broker portal enables the cloud-service-providers to specify available their services. In addition, the cloud-service-consumers may find the most suitable services by negotiating the agreements on the services. The cloud-service-broker as an emerging technology intermediates heterogeneous multiple cloud services for both the providers and consumers. In this paper, we suggest the web-based user interface design of the cloud-service-broker portal to support different providers and consumers.

Design and implementation of zero-copy data path for efficient file transmission

Huge requests for file transmission by concurrent users cause excessive memory copy operations and system calls between kernel and user space, which are very expensive and require many CPU cycles for processing. And it has been a bottleneck that limits the number of serviceable requests and prevents CPU resources from being allocated to other processes. In this paper, we suppose the zero-copy data path for efficient file transmission to solve the upper described problems. To do that, we used existing system call interface, sendfile, for file transmission from one device to another and used the same buffer page to transmit a file. The supposed zero-copy data path reduces CPU cost per transaction by eliminating CPU copies between kernel and user space and by avoiding unnecessary system calls in user space. The improved CPU efficiency means that a larger number of requests can be serviced with the same CPU configuration.

Design and implementation of kernel S/W for TCP/IP offload engine(TOE)

TCP/IP is the most commonly used protocol to communicate between servers, and is used in a wide range of applications. Unfortunately, TCP/IP places a very heavy burden on host CPUs. Server CPUs have begun to choke while processing the TCP/IP overhead associated with transferring data. We have known that the CPU power in todays servers is just sufficient to handle the TCP processing, leaving no CPU cycles for application processing. So, in this paper, to resolve upper problem, we present the design of kernel software for TCP/IP off-loading engine. In this paper, we define the design concept of TOE module, and describe its architecture, main functions and interfaces with TOE device. TOE module consists of three software layers, that is, TOE device driver, TOE offload protocol layer, TOE socket switching layer. TOE device and TOE kernel software have several advantages as below. First, it improves network and application performance, that is, increases network throughput and lowers client response time. Second, it extends the life of existing servers, that is, enabling existing servers to manage additional client loads and enables existing servers to respond more quickly to client queries

Provisioning Virtual Infrastructure to Cloud Service Consumer in Perspective of Brokering VM Placement

From cloud computing environment, cloud service consumers may request virtual infrastructure resources and cloud service providers may provision the virtual resources e.g. virtual machines based on the diverse provisioning options. Through the cloud service broker, the cloud service consumers acquire their on-demand services instead of contacting to each cloud service provider. In order to provide VM services in on-demand infrastructure clouds, we deal with the requirements of each cloud stakeholders and metadata from the view of cloud service broker.

CORSET: Service-Oriented Resource Management System in Linux

Generally, system resources are not enough for many running services and applications in a system. And services are more important than single process in real world and they have different priority or importance. So each service should be treated with discrimination in aspect of system resources. But administrator can’t guarantee the specific service has proper resources in unsettled workload situation because many processes are in race condition. So, we suppose the service-oriented resource management subsystem to resolve upper problems. It guarantees the performance or QoS of the specific service in changeable workload situation by satisfying the minimum resource requirement for the service.

Design of system management architecture using remote storage

As IT resources have increased continually, it is common to adapt management system for resource management. For system management, management server should manage many other systems. Under this situation, system management sewer has several problems, for examples, sewers workload, network bandwidth, storage capacity, management information backup and etc. In this paper, we present system management architecture using remote storage to solve upper problems. In suggested architecture; SM(System Management) agent recognize allocated remote storage as its local storage. And it gathers management information from various system resources, save it new mounted local disk, that is, remote storage. Proposed architecture reduces workload in management sewer. So, it can accommodate increasing IT resources without large cost. Also, it reduces network bandwidth usage and speed up transmission through block level communication.