An operating system which is particularly adapted to heterogenous distributed systems. Entities available to a process running in the operating system are provided by services. Each service models its entity as a set of files. The entity is controlled by performing operations on the set of files provided by the entity. Services other than those provided by the operating system all employ the same protocol specifying operations on files. The only requirement placed on a service by the operating system is that it be able to accept and respond to messages employing the protocol. The files are named, and the operating system provides a plurality of name spaces, relates each process to one of the name spaces, and permits any process to modify its name space or to create a new name space. Services are disclosed which provide each process using the service with its own set of the service's files and which provide files belonging to one process's namespace to another process executing on a different processor.
CROSS REFERENCE TO RELATED APPLICATION
This application is a continuation of application Ser. No. 08/237,967, filed on May 3, 1994, abandoned, which is a continuation of application Ser. No. 07/702,651, filed on May 17, 1991, now abandoned, which is a continuation-in-part of Ser. No. 07/551,218 filed on Jul. 11, 1990 abandoned
The present invention is suitable for use in a communications system that downloads applications over a broadcast file system (BFS). An application services interface (ASI) provides the BFS an application file including a configuration file and an image file that are stored in a service manager in set-tops. Additionally, the ASI provides the BFS a content file that is included in the application file. The provided content file is separated into service files, where each service file is associated with a particular service. When a set-top requests a service, the associated service file is subsequently loaded onto the set-top. Alternatively, a service file is stored in memory until such time as another service is requested, which at that point the older service file is removed from memory to allow space for the newly requested service file.
A file system is distributed over a plurality of computers connected to each other by a network. The plurality of computers execute user programs, and the user programs access files stored on a plurality of physical disks connected to the plurality of computers. The file system includes a plurality of file servers executing on the plurality of computers as a single distributed file server layer. In addition, the file system includes a plurality of disk servers executing on the plurality of computers as a single distributed disk server layer, and a plurality of lock servers executing on the plurality of computers as a single distributed lock server to coordinate the operation of the distributed file and disk server layers so that the user programs can coherently access the files on the plurality of physical disks. The plurality of file servers executes independently on a different one of the plurality of computers, and the plurality of file servers communicate only with plurality of disk servers and the plurality of lock servers, and not with each other. Furthermore, the disk server layer organizes the plurality of physical disks as a single virtual disk having a single address space for the files.
Method and apparatus for atomic file look-up. An atomic look-up operation allows an application to find out whether a file is opened atomically based on whether or not the file path is present in a file system namespace cache. If not, the file open request can be redirected, avoiding or minimizing impacts to the scheduling of various operations involved in executing an application. The request can be redirected by the application to a process that includes blocking point handling. An operating system according to the present invention includes a file system including a file system namespace, and an operating system kernel is operatively connected to the file system. The operating system kernel includes the file system namespace cache and the atomic look-up operation.
A distributed computing system executes application programs which call functions, each function carrying out an operation such as read or write accessing a data file in response to an input argument generated by the calling application program. The system includes client computers for executing the application programs and server computers for executing "server programs", programs which carry out the called functions. The client and server computers communicate through a network. To make a function call, an application program running on a client computer generates an argument for the function and sends a request identifying the function to a "client gate" program also running on the client program. Each client computer stores a configuration file indicating which server star carries out each function, and which server computer can execute that server star. When the client gate receives the function request, it consults the configuration file and then sends a request referencing the identified server star and function to a "server gate" program executed by the identified server. The server gate thereupon invokes the server star which carries out the function. The server gate forwards function output data produced by the server star to the client gate which returns the data to the application program. Thus a change in the network location or structure of a data file used by application programs can be accommodated by modifying configuration files and/or the server stars without necessarily altering the application programs themselves.
Apparatus and methods, including computer program products, for providing a network-accessible queue. The queue is stored on a network-accessible file system having a locking mechanism providing strictly-enforced, network-wide file locking. The data structures of the queue are stored in the file system. The file system is preferably implemented on a network accessible RAID array storage device providing a native CIFS file system for storing queue entries received over a network from sender clients and for delivery of such queue entries to receiver clients. Synchronization is implemented using file locks enforced by the file system, so there is no need for a master synchronization process. Notification of the availability of queue entries to receiver clients is also done by a decentralized process.
