Transparent Fault Tolerance of Device Drivers for Virtual Machines

Heeseung Jo,Hwanju Kim,Jae-Wan Jang,Joonwon Lee,Seungryoul Maeng IEEE 2010 IEEE Transactions on Computers Vol.59 No.11

<P>In a consolidated server system using virtualization, physical device accesses from guest virtual machines (VMs) need to be coordinated. In this environment, a separate driver VM is usually assigned to this task to enhance reliability and to reuse existing device drivers. This driver VM needs to be highly reliable, since it handles all the I/O requests. This paper describes a mechanism to detect and recover the driver VM from faults to enhance the reliability of the whole system. The proposed mechanism is transparent in that guest VMs cannot recognize the fault and the driver VM can recover and continue its I/O operations. Our mechanism provides a progress monitoring-based fault detection that is isolated from fault contamination with low monitoring overhead. When a fault occurs, the system recovers by switching the faulted driver VM to another one. The recovery is performed without service disconnection or data loss and with negligible delay by fully exploiting the I/O structure of the virtualized system.</P>

FAB: flash-aware buffer management policy for portable media players

Jo, Heeseung,Kang, Jeong-Uk,Park, Seon-Yeong,Kim, Jin-Soo,Lee, Joonwon IEEE 2006 IEEE transactions on consumer electronics Vol.52 No.2

This paper presents a novel buffer management scheme for portable media players equipped with flash memory. Though flash memory has various advantages over magnetic disks such as small and lightweight form factor, solid-state reliability, low power consumption, and shock resistance, its physical characteristics imposes several limitations. Most notably, it takes relatively long time to write data in flash memory and the data cannot be overwritten before being erased first. Since an erase operation is performed as a unit of larger block, the employed strategy for mapping logical blocks onto physical pages affects real performance of flash memory. This article suggests a flash-aware buffer management scheme that reduces the number of erase operations by selecting a victim based on its page utilization rather than based on the traditional LRU policy. Our scheme effectively minimizes the number of write and erase operations in flash memory, reducing the total execution time by 17% compared to the LRU policy.

Offloading data encryption to GPU in database systems

Jo, Heeseung,Hong, Seung-Tae,Chang, Jae-Woo,Choi, Dong Hoon Springer-Verlag 2014 The Journal of supercomputing Vol.69 No.1

Improving the startup time of digital TV

Heeseung Jo,Hwanju Kim,Hyun-Gul Roh,Joonwon Lee IEEE 2009 IEEE transactions on consumer electronics Vol.55 No.2

<P>Digital TV was introduced with promises for higher quality displays and converging Internet services which require more computing power and complex software systems. Such complexity entails a prolonged startup time that is similar to booting a computer system. In this paper, we propose two approaches to reduce the startup time of digital TV. Based on the suspend-resume technique employed in many commodity operating systems, our mechanisms replace the cold startup process of digital TV with the resuming from a suspend image. We also analyze the bottlenecks in the startup process and suggest hardware supports to improve the startup time. The proposed approaches reduce the startup time of digital TV by about 50%.</P>

범용 서버 프로세서에서의 FLsched 스케줄러 성능 고찰

조희승(Heeseung Jo) 한국정보과학회 2020 정보과학회 컴퓨팅의 실제 논문지 Vol.26 No.9

프로세서 구조가 매니코어 형태로 발전하고 있고, 한 프로세서에 집적되는 코어의 수는 지속적으로 증가하고 있다. 급속히 증가하는 인공지능과 딥러닝 워크로드는 급속히 증가하고 있고, 계산 집약적이고 매우 큰 병렬성을 요구하는 워크로드 또한 이러한 추세를 가중시키고 있다. 반면, 운영체제에서 매니코어를 지원하기 위한 스케줄러의 발전은 하드웨어 발전 속도를 따라가지 못하고 있지만, 최근 리눅스 운영체제에서 매니코어 프로세서를 고려하여 최적화된 FLsched가 제안된 바 있다. FLsched는 Xeon Phi 프로세서를 고려하여, lockless 디자인과 context switch의 수를 최소화함으로써 성능의 향상을 이루고자 한다. 본 연구에서는 FLsched가 범용 서버급 프로세서에서도 타당성을 가지는지, 성능 향상이 가능한지 분석하고 고찰해 본다. 다수의 서버급 워크로드와 벤치마크를 이용한 성능평가 결과, CFS 대비하여 FLsched가 우수한 성능을 나타내었으며, Xeon Phi에서 보다 그 성능의 향상 정도가 더 큰 것으로 나타났다. 주요한 이유로 FLsched의 lockless 디자인이 주요한 것으로 분석된다. The processor structure has evolved into a manycore processor, and the number of cores integrated into a processor has been continuously increasing. Rapidly increasing AI and deep learning workloads, that are computation-intensive and highly parallel, are also accelerating this trend. Conversely, the development of the OS scheduler to support manycore processors cannot match with the speed of hardware development. But recently, FLsched has been proposed which considers manycore processors in the Linux operating system. FLsched aims the Xeon Phi processor and focuses on improving the scheduler performance by removing the internal locks and by minimizing the number of context switches. In this paper, we analyze and study if FLsched is feasible in general-purpose server-class processors and if the performance can be improved. As a result of a performance evaluation using many server-class workloads and benchmarks, FLsched showed superior performance compared to CFS, and the degree of improvement in the performance was greater than in the Xeon Phi. The lockless design of FLsched is the main reason.

점도 보정을 통한 사출성형CAE 신뢰성 향상에 관한 연구

양희승(Heeseung Yang),조현권(Hyunkwon Jo),박현민(Hyunmin Park),나형현(Hyunghyun Na),박은정(Eunjung Park) 한국자동차공학회 2011 한국자동차공학회 학술대회 및 전시회 Vol.2011 No.11

Recently a quality improvement for car parts is needed in the automotive industry to correspond with the discerning customer’s demands. Especially domestic automotive suppliers are trying to improve the visual quality of car parts because Automotive Interior component made of the plastic is sensitive to visual quality. The plastic injection molding CAE is effective for quality improvement and cost-reduction of mold. So the study on the Reliability Enhancement for CAE is being done. The material properties are a major factor in injection molding CAE and the viscosity is very important factor of them to be considered . But it is realistically impossible to measure viscosity of all material that continually being developed. This paper is purposed to describes the methodology of reliability enhancement for CAE through modification of viscosity.

Anesthetic Management during Electroconvulsive Therapy

You Bin Jo,Heeseung Lee,Soo In Kim,Dong Yeon Kim,Hee Jung Baik,Rack Kyung Chung 이화여자대학교 의과학연구소 2020 EMJ (Ewha medical journal) Vol.43 No.3

Some patients with depression require electroconvulsive therapy performed under general anesthesia. This requires rapid loss of consciousness, with minimal interference from seizure activity, appropriate muscle relaxation status, prevention of hyperdynamic responses to the electrical stimulus, and proper recovery of spontaneous ventilation and consciousness. The authors report the first case of electroconvulsive therapy performed with the patient under general anesthesia at the Ewha Womans University Medical Center, Korea.

Age face simulation using aging functions on global and local features with residual images

Choi, Sung Eun,Jo, Jaeik,Lee, Sanghak,Choi, Heeseung,Kim, Ig-Jae,Kim, Jaihie Elsevier 2017 expert systems with applications Vol.80 No.-

<P><B>Abstract</B></P> <P>In many previous methods for facial age simulation, the shape and appearance features of Active Appearance Models (AAM) are widely used to model the global facial characteristics. However, they cannot sufficiently represent facial details such as spots, scars, fine wrinkles, and skin blemishes, because many of them are removed from the AAM features during the dimension reduction process. Therefore, previous methods are not suitable for real-world applications, such as forensics, face recognition, and entertainment production systems, which require more accurate and realistic age simulation. To overcome the limitation, this paper proposes an automatic age simulation method based on a synergetic combination of the residual image, local features, and AAM global features. The residual image, which is the difference between a facial image and its reconstructed image by using AAM features, contains facial details of the input image that are not included in the AAM features. Representation of facial details in the age simulation process is achieved by generating and adding a targeted age-weighted residual image to the facial image synthesized by the AAM features. Further, facial details such as wrinkles and skin blemishes that have not yet appeared at the current age but normally appear as aging proceeds, are supplemented, and represented by local features that show locally different aging characteristics. The experimental results show that the proposed method simulates a face more accurately and realistically than previous methods, thereby confirming that it is more suitable for the real-world applications.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Age simulation method using global, local features and residual image is proposed. </LI> <LI> The residual image represents the personal characteristics of the input image. </LI> <LI> Local features represent the local aging characteristics of wrinkle and skin. </LI> </UL> </P>

매니코어 스케줄러 연구

이현명(Hyeonmyeong Lee),조희승(Heeseung Jo),이용섭(Yongseob Lee) 한국정보과학회 2021 정보과학회지 Vol.39 No.11

자동차 도어트림 사출성형해석 신뢰도 증진에 대한 실험적 고찰

박재민(Jeamin Park),양희승(Heeseung Yang),조현권(Hyunkwon Jo) 한국자동차공학회 2013 한국자동차공학회 부문종합 학술대회 Vol.2013 No.5