File systems being used as standard software in today's computer syste
ms have to deal with access conflicts when several processes request I
/O operations to the same file. There have to be coordination mechanis
ms for resolving such conflicts according to the prescriptions of the
POSIX standard. Access coordination is usually implemented by executin
g file operations by one process exclusively, sacrificing scalability
of multiple concurrent accesses in multicomputers. To resolve this pro
blem, we propose a scalable coordination scheme which avoids sequentia
lization of accesses by locking only parts of the file (i.e. arbitrary
records). It allows several processes to operate on the same file tru
ly in parallel, while POSIX semantics is retained. Scalability is achi
eved by employing scalable shared memory access mechanisms and scalabl
e elementary locks. The locking scheme has successfully been tested wi
thin an experimental file system implementation.