IBM Hub/Switch Installation Manual

Also see for High Performance Storage System HPSS: Management guide Installation guide

Contents

Chapter 2 HPSS PlanningHPSS Installation Guide September 2002 61Release 4.5, Revision 2that an implementation is thread-safe provided only one thread makes MPI calls. With HPSS MPI-IO, multiple threads will make MPI calls. HPSS MPI-IO attempts to impose thread-safety on thesehosts by utilizing a global lock that must be acquired in order to make an MPI call. However, thereare known problems with this approach, and the bottom line is that until these hosts provide truethread-safety, the potential for deadlock within an MPI application will exist when using HPSSMPI-IO in conjunction with other MPI operations. See the HPSS Programmers Reference Guide,Volume 1, Release 4.5 for more details.Files read and written through the HPSS MPI-IO can also be accessed through the HPSS Client API,FTP, Parallel FTP, or NFS interfaces. So even though the MPI-IO subsystem does not offer all themigration, purging, and caching operations that are available in HPSS, parallel applications canstill do these tasks through the HPSS Client API or other HPSS interfaces.The details of the MPI-IO API are described in the HPSS Programmer’s Reference Guide, Volume 1.2.5.7 DFSDFS is offered by the Open Software Foundation (now the Open Group) as part of DCE. DFS is adistributed file system that allows users to access files using normal Unix utilities and system calls,regardless of the file’s location. This transparency is one of the major attractions of DFS. Theadvantage of DFS over NFS is that it provides greater security and allows files to be shared globallybetween many sites using a common name space.HPSS provides two options for controlling how DFS files are managed by HPSS: archived andmirrored. The archived option gives users the impression of having an infinitely large DFS filesystem that performs at near-native DFS speeds. This option is well suited to sites with largenumbers of small files. However, when using this option, the files can only be accessed through DFSinterfaces and cannot be accessed with HPSS utilities, such as parallel FTP. Therefore, theperformance for data transfers is limited to DFS speeds.The mirrored option gives users the impression of having a single, common (mirrored) name spacewhere objects have the same path names in DFS and HPSS. With this option, large files can be storedquickly on HPSS, then analyzed at a more leisurely pace from DFS. On the other hand, someoperations, such as file creates, perform slower when this option is used, as compared to when thearchived option is used.HPSS and DFS define disk partitions differently from one another. In HPSS, the option for how filesare mirrored or archived is associated with a fileset. Recall that in DFS, multiple filesets may resideon a single aggregate. However, the XDSM implementation provided in DFS generates events on aper-aggregate basis. Therefore, in DFS this option applies to all filesets on a given aggregate.To use the DFS/HPSS interface on an aggregate, the aggregate must be on a processor that hasTransarc’s DFS SMT kernel extensions installed. These extensions are available for Sun Solaris andIBM AIX platforms. Once an aggregate has been set up, end users can access filesets on theaggregate from any machine that supports DFS client software, including PCs. The wait/retry logicin DFS client software was modified to account for potential delays caused by staging data fromHPSS. Using a DFS client without this change may result in long delays for some IO requests.HPSS servers and DFS both use Encina as part of their infrastructure. Since the DFS and HPSSrelease cycles to support the latest version of Encina may differ significantly, running the DFSserver on a different machine from the HPSS servers is recommended.