The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Exibições de texto completo
114
À medida que os sistemas de TI, incluindo os sistemas de rede que utilizam tecnologias SDN/NFV, se tornam em grande escala e complicados, o custo da gestão do sistema também aumenta rapidamente. Os operadores de rede têm de manter o seu fluxo de trabalho na construção e atualização consistente de tais sistemas complexos e, portanto, deseja-se que estas tarefas de gestão na geração do plano de atualização do sistema sejam automatizadas. A atualização declarativa do sistema com busca em espaço de estados é uma abordagem promissora para possibilitar essa automação, porém, os métodos atuais não são suficientemente escaláveis para sistemas práticos. Neste artigo, propomos uma nova abordagem heurística para reduzir significativamente o tempo de computação para resolver o procedimento de atualização do sistema para sistemas práticos. Nossa heurística leva em conta os gargalos estruturais da atualização do sistema e da pesquisa avançada para resolver os gargalos dos estados atuais do sistema. Este artigo inclui as seguintes contribuições: (1) definição formal de uma nova função heurística especializada em atualização de sistema para o algoritmo de busca A*, (2) provas de que nossa função heurística é consistente, ou seja, o algoritmo A* com nossa heurística retorna um ótimo correto solução e pode omitir a expansão repetida de nós em espaços de busca, e (3) resultados de avaliação de desempenho de nossas heurísticas. Avaliamos o algoritmo proposto em dois casos; atualização do hipervisor em execução e atualização contínua de VMs em execução. Os resultados mostram que o tempo de cálculo para resolver o plano de atualização do sistema para um sistema com 100 VMs não excede vários minutos, enquanto o algoritmo convencional só é aplicável para um sistema muito pequeno.
Takuya KUWAHARA
NEC
Takayuki KURODA
NEC
Manabu NAKANOYA
NEC
Yutaka YAKUWA
NEC
Hideyuki SHIMONISHI
NEC
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Takuya KUWAHARA, Takayuki KURODA, Manabu NAKANOYA, Yutaka YAKUWA, Hideyuki SHIMONISHI, "Scalable State Space Search with Structural-Bottleneck Heuristics for Declarative IT System Update Automation" in IEICE TRANSACTIONS on Communications,
vol. E102-B, no. 3, pp. 439-451, March 2019, doi: 10.1587/transcom.2018NVP0009.
Abstract: As IT systems, including network systems using SDN/NFV technologies, become large-scaled and complicated, the cost of system management also increases rapidly. Network operators have to maintain their workflow in constructing and consistently updating such complex systems, and thus these management tasks in generating system update plan are desired to be automated. Declarative system update with state space search is a promising approach to enable this automation, however, the current methods is not enough scalable to practical systems. In this paper, we propose a novel heuristic approach to greatly reduce computation time to solve system update procedure for practical systems. Our heuristics accounts for structural bottleneck of the system update and advance search to resolve bottlenecks of current system states. This paper includes the following contributions: (1) formal definition of a novel heuristic function specialized to system update for A* search algorithm, (2) proofs that our heuristic function is consistent, i.e., A* algorithm with our heuristics returns a correct optimal solution and can omit repeatedly expansion of nodes in search spaces, and (3) results of performance evaluation of our heuristics. We evaluate the proposed algorithm in two cases; upgrading running hypervisor and rolling update of running VMs. The results show that computation time to solve system update plan for a system with 100 VMs does not exceed several minutes, whereas the conventional algorithm is only applicable for a very small system.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2018NVP0009/_p
Copiar
@ARTICLE{e102-b_3_439,
author={Takuya KUWAHARA, Takayuki KURODA, Manabu NAKANOYA, Yutaka YAKUWA, Hideyuki SHIMONISHI, },
journal={IEICE TRANSACTIONS on Communications},
title={Scalable State Space Search with Structural-Bottleneck Heuristics for Declarative IT System Update Automation},
year={2019},
volume={E102-B},
number={3},
pages={439-451},
abstract={As IT systems, including network systems using SDN/NFV technologies, become large-scaled and complicated, the cost of system management also increases rapidly. Network operators have to maintain their workflow in constructing and consistently updating such complex systems, and thus these management tasks in generating system update plan are desired to be automated. Declarative system update with state space search is a promising approach to enable this automation, however, the current methods is not enough scalable to practical systems. In this paper, we propose a novel heuristic approach to greatly reduce computation time to solve system update procedure for practical systems. Our heuristics accounts for structural bottleneck of the system update and advance search to resolve bottlenecks of current system states. This paper includes the following contributions: (1) formal definition of a novel heuristic function specialized to system update for A* search algorithm, (2) proofs that our heuristic function is consistent, i.e., A* algorithm with our heuristics returns a correct optimal solution and can omit repeatedly expansion of nodes in search spaces, and (3) results of performance evaluation of our heuristics. We evaluate the proposed algorithm in two cases; upgrading running hypervisor and rolling update of running VMs. The results show that computation time to solve system update plan for a system with 100 VMs does not exceed several minutes, whereas the conventional algorithm is only applicable for a very small system.},
keywords={},
doi={10.1587/transcom.2018NVP0009},
ISSN={1745-1345},
month={March},}
Copiar
TY - JOUR
TI - Scalable State Space Search with Structural-Bottleneck Heuristics for Declarative IT System Update Automation
T2 - IEICE TRANSACTIONS on Communications
SP - 439
EP - 451
AU - Takuya KUWAHARA
AU - Takayuki KURODA
AU - Manabu NAKANOYA
AU - Yutaka YAKUWA
AU - Hideyuki SHIMONISHI
PY - 2019
DO - 10.1587/transcom.2018NVP0009
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E102-B
IS - 3
JA - IEICE TRANSACTIONS on Communications
Y1 - March 2019
AB - As IT systems, including network systems using SDN/NFV technologies, become large-scaled and complicated, the cost of system management also increases rapidly. Network operators have to maintain their workflow in constructing and consistently updating such complex systems, and thus these management tasks in generating system update plan are desired to be automated. Declarative system update with state space search is a promising approach to enable this automation, however, the current methods is not enough scalable to practical systems. In this paper, we propose a novel heuristic approach to greatly reduce computation time to solve system update procedure for practical systems. Our heuristics accounts for structural bottleneck of the system update and advance search to resolve bottlenecks of current system states. This paper includes the following contributions: (1) formal definition of a novel heuristic function specialized to system update for A* search algorithm, (2) proofs that our heuristic function is consistent, i.e., A* algorithm with our heuristics returns a correct optimal solution and can omit repeatedly expansion of nodes in search spaces, and (3) results of performance evaluation of our heuristics. We evaluate the proposed algorithm in two cases; upgrading running hypervisor and rolling update of running VMs. The results show that computation time to solve system update plan for a system with 100 VMs does not exceed several minutes, whereas the conventional algorithm is only applicable for a very small system.
ER -