Abstract: A framework and method are disclosed for providing/consuming data items. A memory structure is created within an address space allocated to a provider application. The memory structure stores performance counter data generated by the provider application. The counter provider application registers a counter provider and an associated performance counter memory structure via a set of application program interface methods. Thereafter, a counter provider consumer application, via additional application program interface methods, determines the identity of the performance counter provider, sets up a buffer to receive the counter data of the counter provider, and requests extraction of the counter data stored within the performance counter memory structure. The transfer of data from the provider application to the consumer application occurs through the system API.

Abstract: A framework and method are disclosed for carrying out request completion tracing across request identification boundaries. When a request traverses an identification boundary, an event identifies the old and new request identifications and stores this linking information in a trace event record. Thereafter, a consumer of records maintains track of request completion by incorporating the linking information to switch identifications used to track completion of the request across request identification boundaries encountered by the system while responding to the received request. A request trace event record consumer architecture with enhanced tracking capabilities based upon a specified state machine provides greater power when interpreting the trace records associated with completion of requests by a server system.

Abstract: A method of identifying, correcting and displaying application website and device compatibility issues is disclosed. An agent framework is started that host compatibility agents. The compatibility agents are used to analyze an operating system user such as a device, an application or an update. The results of the compatibility agents are communicated to a database and the results of the compatibility agents are displayed using a user interface.

Abstract: Programmatically diagnosing the root cause of a problem in a computing system. Events are monitored within an operating system, and at least a subset of the events are logged to a log file. In response to the detection of error condition(s), a diagnostics module is invoked. The diagnostics module queries the log file to correlate events relevant to diagnosis of the problem, and identifies the root cause by evaluating the results of the query. Once the root cause of the problem is diagnosed, a resolution module corresponding to that root cause may be invoked to programmatically resolve the problem. The invocation of the diagnostic and resolution modules may be subject to policy rules. Furthermore, the logging, diagnostics and resolution modules may be automatically updated as needed.

Abstract: Programmatically diagnosing the root cause of a problem in a computing system. Events are monitored within an operating system, and at least a subset of the events are logged to a log file. In response to the detection of error condition(s), a diagnostics module is invoked. The diagnostics module queries the log file to correlate events relevant to diagnosis of the problem, and identifies the root cause by evaluating the results of the query. Once the root cause of the problem is diagnosed, a resolution module corresponding to that root cause may be invoked to programmatically resolve the problem. The invocation of the diagnostic and resolution modules may be subject to policy rules. Furthermore, the logging, diagnostics and resolution modules may be automatically updated as needed.

Abstract: An event tracing program generally receives performance data about an event occurring on the computer system from a data producer program. The event tracing program responds by recording the event performance data in one of a set of a log buffers. When a log buffer becomes full, the event tracing program places the log buffer on a buffer flush list. The filled buffer is then written out to a more permanent storage medium, such as a disk. From time to time, the event tracing program may also transfer a buffer to the flush list prior to becoming full after a time-out period. To prevent a buffer from being flushed while event performance data is being recorded in the buffer, a reference count is incremented prior to the record operation to signify that the buffer is currently being modified. For high performance on multiprocessor systems, the buffers are allocated per processor to minimize data sharing among processors.

Abstract: A method is provided for estimating the size of cache memory required for optimal performance of a network database service, such as a directory service, by means of an iterative process. In the estimation process, the memory size N for best-case performance (i.e., the memory size that avoids any disk I/O operation) is first determined. The allocated memory size is then given a starting value. The probability (p) of cache-miss is then estimated for that memory size. Another probability (q), which is the probability that a record requested by a frequent user of the service is not in the cache, is also estimated for the memory size. The performance impact of the disk I/O rate indicated by p and q is then evaluated. If the performance is not adequate, the cache memory size is adjusted to a different value. The miss probabilities p and q are again estimated, and the performance impact is estimated. This iterative process is continued until the cache memory size is found to provide adequate estimated performance.