• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.127-148
• /
• 2020

The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.

• Korean Journal of Agricultural Science
• /
• v.10 no.1
• /
• pp.28-60
• /
• 1983

This study was carried out to define the effects of external factors including temperature, moisture, oxygen and light quality on the germination of sesame seeds and to investigate the change of major chemical constituents of seeds during germination. The results obtained are summarized as follows: 1. The average germination ratio was from 95.8% to 97.2% when it was tested every $5^{\circ}C$ intervals from $20^{\circ}C$ to $35^{\circ}C$ and no significant difference in germination ratio was found within $20^{\circ}C$ to $35^{\circ}C$. But the germination ratio dropped rapidly to 32.2% when seeds were germinated at $15^{\circ}C$ and the coefficient of variation become greater(77%) 2. The days required for germination ranged from 1.16 to 1. 64 at the temperatures of $35^{\circ}C$ to $25^{\circ}C$ and they were 3.07 and 10.4 at the temperatures of $20^{\circ}C$ and $15^{\circ}C$, respectively. 3. Considering the germination ratio and days needed, $15^{\circ}C$ was assumed to be the minimum temperature for germination practically and this temperature is recommended for testing low temperature tolerance of seed germination of sesame cultivars. 4. The varieties shown the highest low temperature tolerance were Shirogoma and Turkey. The next varieties shown some degree of low temperature germination were Suweon #29, Naebok and IS 58. The varieties with 70 to 80% of germination ratio were Maepo, Suweon #14, Kimpo, Moondeok, and Haenam. Among the 90 varieties tested, the varieties with comparatively high degree of low temperature tolerance were about 10%, and 70% of the low temperature tolerant varieties were domestic varieties. 5. At $12^{\circ}C$ the Shirogoma was the only variety which showed over 50% of germination ratio, 71.4% of the varieties showed less than 20% of germination ratio. When the temperature was raised to $27^{\circ}C$ 18 days after placement at $12^{\circ}C$ all the varieties showed over 90% of germination ratio within 2days. 6. The amounts of water imbibition needed for seed germination were 0.48 to 0.62 times of the seed dry weight at $25^{\circ}C$ and were significantly different among sesame cultivars. About 63% of water required for germination was imbibed in 2 hours after placement of seeds under the germination condition. 7. Under saturated moisture condition the average germination ratio was 0.42%. In the soil of which water potential was -0.4bar 64.8% of the seeds germinated and the most adequate soil water potential for sesame seed germination was about -0.4 to -5.5 bar. The germination ratio decreased as the soil water potential declined below -5.5 bar. 8. Six out of 10 varieties were not influenced by 5% of oxygen in air germination chamber, while varieties such as Yecheon, PI 158073, IS 103 and Euisangcheon showed 64 to 91% of germination under the 5% oxygen content. Under anaerobic condition, cotyledones were not emerged but only hypocotyl was emerged and elongated. The germination ratio of IS 103 decreased significantly under anaerobic condition. 9. When the seeds were dried for 24 hours after 12 hours imbibition of water, the seeds of Cheongsong did not lose their germination ability and 27.5% was germinated but Suweon #9 and Early Russian failed to germinate. However, the germination ratio of IS 103 decreased when the seed were dried 24 hours after 4 hours imbibition of water and the germination ability of IS 103 was maintained even though the seeds were dried for 24 hours after 24 hours imbibition of water. 10. During germination, sugar content of sesame seed increased rapidly and activity of ${\alpha}$-amylase increased gradually while starch content decreased significantly. The rates of increase in sugar content and enzyme activity and decrease in starch content were significantly lower at $15^{\circ}C$ compared with those at $25^{\circ}C$. 11. During germination of sesame seeds, lipid content in the seeds dropped rapidly and the activity of alkaline lipase increased significantly at early stage of germination. The rate of decrease in lipid content and increase in emzyme activity was lower at $15^{\circ}C$ than at $25^{\circ}C$. 12. Four out of 6 varieties were not affected in germination by light wave length. But Suweon #8 was inhibited in germination by 600-650nm. and IS 103 by 600 to 650nm and 500 to 550nm of light wave length. Suweon #8 showed high germination ratio under 650 to 760 nm and 500 to 560nm, and IS 103 under 400 to 470nm and complete darkness. 13. The germination ratios increased significantly in the seeds of which 1000 grain weight is heavier. When the seeds were placed at soil 4cm deep, Cheongsong and Early Russian failed to emerge their cotyledones, but Suweon #9 and IS 103 showed 32.5 and 50% cotyledone emergence, respectively. The extracts from sesame plant and soil where the sesame was cultivated previously did not affect in the-germination of sesame seeds. 14. The covering by black or transparent polyethylene films increased germination ratio compared with uncovered seeds. The covering was effective in shortening the days needed for germination and in improving the early seedling growth, number of capsules per plant and grain yield. Difference was not so seizable between the two polyethylene films but the transparent film appeared somewhat more effective than the black one. 15. Simcheon, Cheongsong. Suweon #9. PI 158073 and IS 103 showed lower rate of water absorbtion by seed during germination and Suweon #8, Suweon #26, Orotall and Euisangcheon showed high increase in seed weight after water absorbtion by seed.