{"title":"阵列存储系统地址空间溢出的有效维护","authors":"M. Omar, K. Hasan","doi":"10.1109/PDCAT.2016.040","DOIUrl":null,"url":null,"abstract":"Array based storage and retrieval systems are demanded in many high dimensional systems like Big data for their easy maintenance. However, the lack of scalability of the conventional approaches degrades with the dynamic size of data sets as they entail reallocation in order to preserve expanded data velocity. To maintain the velocity of data, the storage system must be scalable enough by allowing subjective expansion on the boundary of array dimension. Again, for an array based storage system, if the number of dimension and length of each dimension of the array is very high then the required address space overflows and hence it is impossible to allocate such a big array in the memory. The index array offers a dynamic storage scheme for preserving expanded data velocity by employing indices for each dimension. In this paper we demonstrate a scalable array storage scheme that divides expanded data size into segments. Hence it is able to maintain overflow and can improve the storage utilization than the conventional one. The system converts the n dimensions of the array into 2 dimensions, hence it involves only 2 indices which ensures lower cost of index computation and higher data locality.","PeriodicalId":203925,"journal":{"name":"2016 17th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Towards an Efficient Maintenance of Address Space Overflow for Array Based Storage System\",\"authors\":\"M. Omar, K. Hasan\",\"doi\":\"10.1109/PDCAT.2016.040\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Array based storage and retrieval systems are demanded in many high dimensional systems like Big data for their easy maintenance. However, the lack of scalability of the conventional approaches degrades with the dynamic size of data sets as they entail reallocation in order to preserve expanded data velocity. To maintain the velocity of data, the storage system must be scalable enough by allowing subjective expansion on the boundary of array dimension. Again, for an array based storage system, if the number of dimension and length of each dimension of the array is very high then the required address space overflows and hence it is impossible to allocate such a big array in the memory. The index array offers a dynamic storage scheme for preserving expanded data velocity by employing indices for each dimension. In this paper we demonstrate a scalable array storage scheme that divides expanded data size into segments. Hence it is able to maintain overflow and can improve the storage utilization than the conventional one. The system converts the n dimensions of the array into 2 dimensions, hence it involves only 2 indices which ensures lower cost of index computation and higher data locality.\",\"PeriodicalId\":203925,\"journal\":{\"name\":\"2016 17th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT)\",\"volume\":\"29 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 17th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/PDCAT.2016.040\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 17th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PDCAT.2016.040","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Towards an Efficient Maintenance of Address Space Overflow for Array Based Storage System
Array based storage and retrieval systems are demanded in many high dimensional systems like Big data for their easy maintenance. However, the lack of scalability of the conventional approaches degrades with the dynamic size of data sets as they entail reallocation in order to preserve expanded data velocity. To maintain the velocity of data, the storage system must be scalable enough by allowing subjective expansion on the boundary of array dimension. Again, for an array based storage system, if the number of dimension and length of each dimension of the array is very high then the required address space overflows and hence it is impossible to allocate such a big array in the memory. The index array offers a dynamic storage scheme for preserving expanded data velocity by employing indices for each dimension. In this paper we demonstrate a scalable array storage scheme that divides expanded data size into segments. Hence it is able to maintain overflow and can improve the storage utilization than the conventional one. The system converts the n dimensions of the array into 2 dimensions, hence it involves only 2 indices which ensures lower cost of index computation and higher data locality.