Further Implementation Of The User Space File System Based On Castorfs And Xrootd

The LHC (Large Hadron Collider) experiments use a mass storage system forrecording petabytes of experimental data. This system is called CASTOR[1](CERNAdvanced STORage manager) and it is powerful and convenient for many use cases.However, it is impossible to use standard tools and scripts straight away for dealingwith the files in the system since the users can access the data just in the way ofusing command-line utilities and parsing their output. Thus a complete POSIXfilesystem–CastorFS[2]is developed based on the FUSE[3](File System inUserspace) and two CASTOR I/O libraries-RFIO (Remote File I/O) library and NS(Name Server) library. Although it is applied successfully, it has serious limitationof wide application because the I/O protocols it relies on are very old. Each time theCASTOR side receives the calling for accessing data files from the user side, theload of the CASTOR side system will increase quickly even out of its upper boundand the system would crash. Besides that, those two protocols provide veryprimitive mechanism for authentication and the data transmitting rate is very low.Furthermore, since the CastorFS is implemented in C language, it can hardly beextended by using the other well functioned libraries. Hence, it needs to berewritten in C++with the implementation of caching and the Xrootd[4](eXtendedRoot Daemon) which is developed by SLAC/CERN to provide a high performance,scalable fault tolerant access to data. The new system implemented the LRU (LeastRecently Used) algorithm as the mechanism to manage the cache. We use FUSE asthe tool to provide the interfaces for the virtual filesystem. By implementing theactual filesystem function with Xrootd Posix APIs, we provided a filesystem whichcan effectively ease the server load, provide a steady data transmitting service andprovide a good data protection mechanism.