• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권10호
• /
• pp.5058-5077
• /
• 2019

Traditional regenerating codes are designed to tolerate node failures with optimal bandwidth overhead. However, there are many types of partial failures inside the node, such as latent sector failures. Recently, proposed regenerating codes can also repair intra-node failures with node-level redundancy but incur significant bandwidth and I/O overhead. In this paper, we construct a new scheme of regenerating codes, called IR-RBT codes, which employs intra-node redundancy to tolerate intra-node failures and serve as the help data for other nodes during the repair operation. We propose 2 algorithms for assigning the intra-node redundancy and RBT-Helpers according to the failure probability of each node, which can flexibly adjust the helping relationship between nodes to address changes in the actual situation. We demonstrate that the IR-RBT codes improve the bandwidth and I/O efficiency during intra-node failure repair over traditional regenerating codes but sacrifice the storage efficiency.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3583-3598
• /
• 2019

In order to improve the reliability and repair efficiency of distributed storage systems, minimum bandwidth regenerating (MBR) codes based on cyclic variable fractional repetition (VFR) codes are constructed in this thesis, which can repair failed nodes accurately. Specifically, in order to consider the imbalance of data accessed by the users, cyclic VFR codes are constructed according to that data with different heat degrees are copied in different repetition degrees. Moreover, we divide the storage nodes into groups, and construct MBR codes based on cyclic VFR codes to improve the file download speed. Performance analysis and simulation results show that, the repair locality of a single node failure is always 2 when MBR codes based on cyclic VFR codes are adopted in distributed storage systems, which is obviously superior to the traditional MBR codes. Compared with RS codes and simple regenerating codes, the proposed MBR codes based on cyclic VFR codes have lower repair locality, repair complexity and bandwidth overhead, as well as higher repair efficiency. Moreover, relative to FR codes, the MBR codes based on cyclic VFR codes can be applicable to more storage systems.

• 한국통신학회논문지
• /
• 제41권12호
• /
• pp.1684-1691
• /
• 2016

최근 분산 저장 시스템에 erasure code를 활용하여 저장소 효율성을 높이려는 연구가 활발히 진행되고 있다. 재생 부호(regenerating codes)는 erasure code의 일종으로, 높은 저장소 효율성과 네트워크 효율성을 가지는 코드이다. (n,k,d)-재생 부호는 n개의 저장소 노드를 가지며, 손실된 노드가 발생하였을 때, 해당 노드는 d개의 살아남은 노드로부터 정보를 다운로드받아 복구될 수 있다. 하지만 일반적인 재생 부호는 노드 복구 시 정확히 d개의 도움 노드들을 사용해야 하며, 노드 손실이 빈번하거나, 노드 간 접속이 불안정한 환경에서, d개 이하의 노드들에만 접속 가능할 경우에 유연하게 대처할 수 없다. 본 논문에서는 약간의 복구 대역폭의 희생을 통하여, $$k{\leq_-}{\bar{d}}{\leq_-}d$$의 다양한 도움 노드의 수 ${\bar{d}}$개로 노드를 복구할 수 있는 유연한 코드 운용 방식을 제안하였다.

• ETRI Journal
• /
• 제38권3호
• /
• pp.529-539
• /
• 2016

Recently, a product-matrix (PM) framework was proposed to construct optimal regenerating codes for homogeneous distributed storage systems (DSSs). In this paper, we propose an extended PM (EPM) framework for coding of heterogeneous DSSs having different repair bandwidths but identical storage capacities. Based on the EPM framework, an explicit construction of minimum remote-repair bandwidth regenerating (MRBR) codes is presented for a specific heterogeneous DSS, where two geographically different datacenters with associated storage nodes are deployed. The data reconstruction and regeneration properties of the MRBR code are proved strictly. For the purpose of demonstration, an example implementation of MRBR code is provided. The presented MRBR code is the first optimal strict-regenerating code for heterogeneous DSSs. In addition, our proposed EPM framework can be applied to homogeneous systems also.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권4호
• /
• pp.1182-1199
• /
• 2023

Fractional repetition (FR) codes can achieve exact uncoded repair for multiple failed nodes, with lower computational complexity and bandwidth overhead, and effectively improve repair performance in distributed storage systems (DSS). The actual distributed storage system is dynamic, that is, the parameters such as node storage overhead and number of storage nodes will change randomly and dynamically. Considering that traditional FR codes cannot be flexibly applied to dynamic distributed storage systems, a new construction scheme of adaptive-and-resolvable FR codes based on hypergraph coloring is proposed in this paper. Specifically, the linear uniform regular hypergraph can be constructed based on the heuristic algorithm of hypergraph coloring proposed in this paper. Then edges and vertices in hypergraph correspond to nodes and coded packets of FR codes respectively, further, FR codes is constructed. According to hypergraph coloring, the FR codes can achieve rapid repair for multiple failed nodes. Further, FR codes based on hypergraph coloring can be generalized to heterogeneous distributed storage systems. Compared with Reed-Solomon (RS) codes, simple regenerating codes (SRC) and locally repairable codes (LRC), adaptive-and-resolvable FR codes have significant advantages over repair locality, repair bandwidth overhead, computational complexity and time overhead during repairing failed nodes.

• 한국통신학회논문지
• /
• 제40권9호
• /
• pp.1741-1753
• /
• 2015

본 논문에서는 MBR 재생부호인 반복분할 부호의 부분접속수를 향상시킬 수 있는 방법에 대해 소개한다. 향상된 부분접속수를 갖는 반복분할 부호를 부분접속 복구 가능한 반복분할 부호라고 한다. 부분접속 복구 가능한 반복분할 부호의 서로 다른 두 가지 생성 방법을 소개하고 각각을 다양한 성능 척도를 통해 분석한다. 새로운 부호는 반복분할 부호에 비해 낮은 부분접속수를 갖는 대신 저장 가능한 최대 파일 크기나 필요한 저장 노드의 수에서 손해가 발생한다. 다른 부분접속 복구 부호와 비교해 향상된 안정성을 갖고 또한 복구시 심벌 연산이 필요 없는 단순전달 복구를 수행함으로써 복구 복잡도를 낮출 수 있다.

• 한국통신학회논문지
• /
• 제39B권11호
• /
• pp.723-729
• /
• 2014

본 논문에서는 임의의 선형 재생 부호 앙상블에 대하여 복구 대역폭(Repair bandwidth)과 접속 비용(Repair read cost)의 평균을 유도한다. 한 데이터가 여러 노드에 부호화 되어 분산 저장된 상황에서 하나의 노드가 소실될 경우, 이를 복구하기 위해 필요한 데이터 량을 복구 대역폭, 접속해야 하는 노드 수를 복구 접속 비용이라 한다. 이 때, 선형 재생 부호 앙상블은 데이터 심볼의 수 k와 패리티 심볼의 수 m, 그리고 그들의 차수 분포로 주어진다. 우리는 이러한 부호들이 시스터메틱(Systematic)이며 정확한 복구(Exact repair)를 수행하고 n=k+m개의 모든 저장소(Storages)들이 전부 연결되어 있는 상황을 가정한다. 본 논문의 결과는 파운틴 부호 등과 같이 위와 같은 파라미터들로 랜덤하게 만들어진 부호들에 바로 적용 가능하다. 최종 결과식은 평균 복구 접속 비용이 차수 분포와 n, k에 따라 결정됨을 보여준다.

• 한국자동차공학회논문집
• /
• 제13권2호
• /
• pp.52-57
• /
• 2005

The increasing automobiles continue to cause air-pollution problem s worse than ever. In fact, many automobile research are involved in how to reduce exhaust emissions effectively specially in $NO_X$ and PM to comply with stringent emission standards, Euro V. This research emphasized on the development of continuous regeneration DPF technology which was one of promising removing technology of particulate matters because of its comparability and high applicability. In addition, this research discussed on some design points of view through correlation study by com paring the experimental data with computational results by the introduction of commercial codes such as CFD-ACE+ and KIVA-3V. The numerical simulation on the performance of continuous regeneration DPF apparatus and corresponding emission characteristics has been predicted well enough and verified with experimental results. The pressure and average temperatures are decreased to about 2.6% and 1.4% respectively under a full engine load condition mainly due to back pressures raised by diesel particulate filter. Pressure, temperature and heat releasing rates tend to decrease specially at higher engine load, but they are not affected at lower engine load regions.