Abstract: A network switch, in response to receipt from a source station of a Layer 2 reservation request, establishes a reservation for capacity of an ingress queue of the network switch for a data flow of the source station. In response to a queue overrun condition on the ingress queue of the network switch while the reservation is active, the network switch preserves data frames in the data flow of the source station transmitted pursuant to the reservation and discards other data frames.

Abstract: A network switch, in response to receipt from a source station of a Layer 2 reservation request, establishes a reservation for capacity of an ingress queue of the network switch for a data flow of the source station. In response to a queue overrun condition on the ingress queue of the network switch while the reservation is active, the network switch preserves data frames in the data flow of the source station transmitted pursuant to the reservation and discards other data frames.

Abstract: A multiprocessor computing system has memory shared by all processors of the computing system and includes an symmetric multiprocessor (SMP) operating system and at least one external device controlled by a device driver. The device driver is typically written to run in a uniprocessor environment. A method for protecting the execution state of the device driver uses a pair of "collaborative" spinlocks, an interrupt time spinlock and a task time spinlock. At task time, prior to calling the device driver, the task time spin lock is acquired by the operating system. The operating system then waits until the interrupt time spinlock is in an "unowned" state before calling the device driver. Upon return from the device driver, the operating system releases the task time spinlock. At interrupt time, prior to calling the device driver's interrupt handler, the interrupt time spinlock is acquired by the operating system. The operating system then determines if the task time spinlock is owned.

Abstract: A file-based Read-ahead method employs asynchronous I/O processes to fetch Demand and Read-ahead data blocks from a disk (4), depending on their physical and logical sequentialities. When a file system (10) issues an I/O address to obtain data blocks from a file, it also determines the disk location of data blocks that are logically consecutive with the demanded data. The number of such data blocks to be determined in this process is accomplished in accordance with a dynamic Read-ahead method. If the data blocks are physically consecutive, a single I/O request is sent to a disk driver to obtain both the Demand data block and to Prefetch the Read-Ahead data block(s). If, instead, the data blocks are found not to be physically consecutive, a plurality of asynchronous disk I/O requests are issued so as to separately obtain the Demand and the Read-Ahead data blocks.

Abstract: Methods and apparatus provide external interrupt serialization compatibility for a multiprocessing environment so that software written to run in a uniprocessor environment can run without degradation caused by inadequate or improper handling. When a processor receives an external interrupt, the interrupt manager will acquire an external interrupt spinlock. Once acquired, the interrupt manager will transfer control to the external interrupt service routine and upon return from that routine, the interrupt manager will release the external interrupt spinlock. Also, protection of processor execution state in a multiprocessor environment is provided using the same spinlock.

Abstract: A file-based Read-ahead method employs asynchronous I/O processes to fetch Demand and Read-ahead data blocks from a disk (4), depending on their physical and logical sequentialities. When a file system (10) issues an I/O address to obtain data blocks from a file, it also determines the disk location of data blocks that are logically consecutive with the demanded data. The number of such data blocks to be determined in this process is accomplished in accordance with a dynamic Read-ahead method. If the data blocks are physically consecutive, a single I/O request is sent to a disk driver to obtain both the Demand data block and to Prefetch the Read-Ahead data block(s). If, instead, the data blocks are found not to be physically consecutive, a plurality of asynchronous disk I/O requests are issued so as to separately obtain the Demand and the Read-Ahead data blocks.

Abstract: Lazy Write disk cache systems are often utilized in disk cache designs. In such systems, data to be written to the disk storage system is temporarily placed in cache memory for a preselected period of time until an actual update of the disk occurs. By deferring disk write operations input/output performance efficiency is enhanced. However, in sector-based disk storage systems in which data may only be read from or written to a disk in predetermined fixed amounts an attempted update of data often requires a read of data from the disk storage system to ensure that existing data is not written over during an update. The method and system of the present invention avoids this problem by maintaining a status table for each block of cache memory which includes an identification of each byte within each block of cache memory which will be updated.

Abstract: A least recently used cache replacement system in which the data cache is logically partitioned into two separate sections, demand and prefetch. A cache directory table and a least recently used table are used to maintain the cache. When a new demand data page is added to the cache, a most recently used (MRU) pointer is updated and points to this new page. When a prefetch page is added to the cache, the least recently used pointer of the demand section is updated with its backward pointer pointing to this new page. A cache hit on a demand of prefetch page moves that page to the top of the least recently used table. When a free page is needed in the cache, it is selected from the demand or prefetch sections of the memory based on a comparison of the demand hit density and the prefetch hit density so to maintain a balance between these two hit densities.

Abstract: A data processing system has a microprocessor, a memory system and a plurality of system devices the operation of which is controlled by BIOS routines. The BIOS routines are accessed through protected entry points. When an application program attempts to access one of the routines by using a hard coded instruction for jumping to such entry point, a BIOS signaling routine is executed which provides a signal to an operating system allowing the operating system to control the access without being bypassed.