It was in the year 2010 that the concept of 'analytic platform' began to cause a stir with its untapped potential. Ever since, there has been no looking back, and this concept continues to spark a flurry of interest among the organizations. However, there are some organizations that are yet to come to terms with the fact that this platform is quite different from a traditional DBMS. There are many other facts that are still unknown to most of the organizations. Some of these facts are as follows:
• The functionality of this platform is not the same as that of an enterprise data warehouse (EDW); however, there are some similarities. For instance, both of these are supposedly used for data management.
• This platform needs to have some basis features, or else it would prove to be of no use. For instance, it should provide some native support for Map Reduce (MR). It is not unknown that Map Reduce acts as an interface, and provides the requisite framework for writing as well as deploying applications.
• There has to be some integration with SQL at any cost. Integration helps in a lot of ways. To begin with, it allows the SQL analysts to make use of the Map Reduce applications. Furthermore, SQL is known to have exceptional flexibility and power. These features cam be easily used by the MR data scientists in order to complement the MR applications.
• The ideal analytic platform should be cost-effective, and easy to use. A typical platform of such kind should be accompanied by data visualization tools, libraries of pre-packaged SQL/MR analytic functions, and other crucial elements. Furthermore, it is noteworthy that a platform which is too expensive would not serve any purpose.
• It is not known to many organizations that the analysis of data is very much required if they need to manage big and diverse data. In other words, the analytic platform goes hand in hand with the data management.
• For effective data management, it is essential that the platform in question is unique. As a matter of fact, the experts suggest that it should be a lot different from even some of the pure MPP DBMS's. Furthermore, it should have a native in-database processing engine that can be used for the purpose of application embedding. It is also essential that the ideal platform offers 100% enterprise feature support for almost each and every interface (read: SQL, Map-reduce ET AL).
• The functionality of this platform is not the same as that of an enterprise data warehouse (EDW); however, there are some similarities. For instance, both of these are supposedly used for data management.
• This platform needs to have some basis features, or else it would prove to be of no use. For instance, it should provide some native support for Map Reduce (MR). It is not unknown that Map Reduce acts as an interface, and provides the requisite framework for writing as well as deploying applications.
• There has to be some integration with SQL at any cost. Integration helps in a lot of ways. To begin with, it allows the SQL analysts to make use of the Map Reduce applications. Furthermore, SQL is known to have exceptional flexibility and power. These features cam be easily used by the MR data scientists in order to complement the MR applications.
• The ideal analytic platform should be cost-effective, and easy to use. A typical platform of such kind should be accompanied by data visualization tools, libraries of pre-packaged SQL/MR analytic functions, and other crucial elements. Furthermore, it is noteworthy that a platform which is too expensive would not serve any purpose.
• It is not known to many organizations that the analysis of data is very much required if they need to manage big and diverse data. In other words, the analytic platform goes hand in hand with the data management.
• For effective data management, it is essential that the platform in question is unique. As a matter of fact, the experts suggest that it should be a lot different from even some of the pure MPP DBMS's. Furthermore, it should have a native in-database processing engine that can be used for the purpose of application embedding. It is also essential that the ideal platform offers 100% enterprise feature support for almost each and every interface (read: SQL, Map-reduce ET AL).
No comments:
Post a Comment