{"title":"全硅数据中心,能源展望","authors":"Antonio Scarfò","doi":"10.1109/WAINA.2013.186","DOIUrl":null,"url":null,"abstract":"The energy consumption related to data centers is ever increasing, despite the significant efficiency improvements in the involved technologies and the economic crisis. Pervasive adoption of virtualization, decreasing of PUE as well as new features related to energy-efficiency are not able to stop this growth trend, mainly sustained from the explosion of cloud services in both the fixed and mobile ICT markets. One of the most important factors that contributes to the increase in the data center energy demand is the evolution of data utilization in cloud environments. Big data support is one of the main driver of this process, new data sources continuously produce huge amounts of mostly unstructured data, that should be stored in large centralized storage systems to be accessed by the cloud users and analyzed by the companies in real time. This has a disruptive effect for the energy consumption of the data center hosting the above storage resources, that's why the data need to be stored and analyzed in IOPS intensive style. Flash memory technologies can play a key role in such scenario, since they can reduce dramatically both the storage footprint and energy consumption and increase applications performances bringing appreciable economic benefits. This fundamental efficiency and performance improvement for storage architectures can't be supported by legacy technologies because they are not designed to use flash technologies but rather they are based on energy-hungry legacy mechanical disks. That is limit the energy efficiency and performance delivered by solid state disks (SSD) installed in legacy storage systems, opening the door to the development of new technologies that will be able to fully exploit the potential of flash disks. The aim of this work is presenting a brief survey about the potential of flash technologies for improving the energy efficiency of data center storage systems supporting the incoming Big Data services.","PeriodicalId":359251,"journal":{"name":"2013 27th International Conference on Advanced Information Networking and Applications Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"All Silicon Data Center, the Energy Perspectives\",\"authors\":\"Antonio Scarfò\",\"doi\":\"10.1109/WAINA.2013.186\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The energy consumption related to data centers is ever increasing, despite the significant efficiency improvements in the involved technologies and the economic crisis. Pervasive adoption of virtualization, decreasing of PUE as well as new features related to energy-efficiency are not able to stop this growth trend, mainly sustained from the explosion of cloud services in both the fixed and mobile ICT markets. One of the most important factors that contributes to the increase in the data center energy demand is the evolution of data utilization in cloud environments. Big data support is one of the main driver of this process, new data sources continuously produce huge amounts of mostly unstructured data, that should be stored in large centralized storage systems to be accessed by the cloud users and analyzed by the companies in real time. This has a disruptive effect for the energy consumption of the data center hosting the above storage resources, that's why the data need to be stored and analyzed in IOPS intensive style. Flash memory technologies can play a key role in such scenario, since they can reduce dramatically both the storage footprint and energy consumption and increase applications performances bringing appreciable economic benefits. This fundamental efficiency and performance improvement for storage architectures can't be supported by legacy technologies because they are not designed to use flash technologies but rather they are based on energy-hungry legacy mechanical disks. That is limit the energy efficiency and performance delivered by solid state disks (SSD) installed in legacy storage systems, opening the door to the development of new technologies that will be able to fully exploit the potential of flash disks. The aim of this work is presenting a brief survey about the potential of flash technologies for improving the energy efficiency of data center storage systems supporting the incoming Big Data services.\",\"PeriodicalId\":359251,\"journal\":{\"name\":\"2013 27th International Conference on Advanced Information Networking and Applications Workshops\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-03-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2013 27th International Conference on Advanced Information Networking and Applications Workshops\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/WAINA.2013.186\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 27th International Conference on Advanced Information Networking and Applications Workshops","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WAINA.2013.186","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
The energy consumption related to data centers is ever increasing, despite the significant efficiency improvements in the involved technologies and the economic crisis. Pervasive adoption of virtualization, decreasing of PUE as well as new features related to energy-efficiency are not able to stop this growth trend, mainly sustained from the explosion of cloud services in both the fixed and mobile ICT markets. One of the most important factors that contributes to the increase in the data center energy demand is the evolution of data utilization in cloud environments. Big data support is one of the main driver of this process, new data sources continuously produce huge amounts of mostly unstructured data, that should be stored in large centralized storage systems to be accessed by the cloud users and analyzed by the companies in real time. This has a disruptive effect for the energy consumption of the data center hosting the above storage resources, that's why the data need to be stored and analyzed in IOPS intensive style. Flash memory technologies can play a key role in such scenario, since they can reduce dramatically both the storage footprint and energy consumption and increase applications performances bringing appreciable economic benefits. This fundamental efficiency and performance improvement for storage architectures can't be supported by legacy technologies because they are not designed to use flash technologies but rather they are based on energy-hungry legacy mechanical disks. That is limit the energy efficiency and performance delivered by solid state disks (SSD) installed in legacy storage systems, opening the door to the development of new technologies that will be able to fully exploit the potential of flash disks. The aim of this work is presenting a brief survey about the potential of flash technologies for improving the energy efficiency of data center storage systems supporting the incoming Big Data services.