Abstract: Load balancing is achieved at the client side, rather than at the server side of a client-server network. Each client computer regularly receives a load balance list, enumerating respective addresses of multiple server computers. Each client computer executes a server selection function which determines the average load for each server in the list. In the event of a server computer failure, a system administrator can remove the server computer from the load balance list and reapportion the load. The client computer's list then is updated when the list is received during subsequent access. In the event a client computer determines that a server is non-responsive, such server is removed from the load balance list for the client computer which made such determination. The client computer also stores a back-up list of servers for use when all servers on the load balance list are non-responsive.

Abstract: Computer system activity is monitored to identify triggers. A trigger is an event or sequence of events which is a reliable predictor of the subsequent occurrence of a specific repeated disk-intensive interval. A disk-intensive interval is an interval of time in which disk read and/or disk write operations occur at a much greater than average rate. Once a trigger and a corresponding repeated disk-intensive interval are identified, a time log constructor is invoked. The time log rearranges the input/output for faster access and loading of the cache. This reduces disk drive latency. When the trigger is detected, instead of waiting for the data request the computer processes the time log to preload the disk data into cache in advance of the data request. When the data request occurs thereafter, the data is already present in the cache.

Abstract: A data pattern cache stores data patterns in units (i.e., a sector). A given data pattern is associated with one or more physical device address ranges which store such data pattern (e.g., sectors). Thus, although the same data pattern may occur as 10 redundant sectors for different files on different parts of a hard disk, the data pattern is stored in the data pattern cache once. One or more physical device sectors are linked to the data pattern.

Abstract: A user views search results and subjectively determines if a document is desirable or undesirable. Only documents categorized by the user are analyzed for deriving a list of prospective keywords. The frequency of occurrence of each word of each document is derived. Keywords that occur only in desirable documents are good keywords. Keywords that occur only in undesirable documents are bad keywords. Keywords that occurs in both types are dirty keywords. The best keywords are the good keywords with the highest frequency of occurrence. The worst keywords are the bad keywords with the highest frequency of occurrence. A new query phrase includes the highest ranked good keywords and performs filtering using the highest ranked bad keywords. Key phrases are derived to clean dirty keywords into good key phrases. A key phrase also is derived from a good keyword and replaces the good keyword to narrow a search.

Abstract: Launch time for a computer program is reduced by logging hard disk accesses during an initial launch, then processing the log file to accelerate subsequent launches. The log file is processed by identifying all the file portions accessed during the launch, eliminating any duplicate cluster accesses, then sorting the remaining accesses. The disk access log entries are sorted by physical address or are grouped by file, then organized by logical address within each group. The processed log file is stored with the application program. When the application program is launched thereafter, the processed log file is accessed first. All the disk accesses in the log file are performed moving all the data into RAM cache. When the program launch resumes, the launch occurs faster because all the data is already in cache.

Abstract: A computer program's launch time is reduced by reorganizing physical address storage of the program on a secondary storage device. Launch of the program is detected, then file access during the launch is monitored. A log file entry is created for each access to the secondary storage device. After launch is complete, the portions of the computer program accesses during the launch are moved into a contiguous area of the secondary storage device. Placement within the area corresponds to the access order during launch. Subsequent launches are faster because the launch portions of the program are contiguously located. Only the first access to a given memory block is considered in ordering the portions within the contiguous area The cluster size is used to estimate whether differing log entries correspond to the same address block.

Abstract: Disk I/O average access time is improved by estimating access time for pending I/O requests and selecting the request with the lowest estimated access time as the next request to be processed. To provide fast run-time estimation, disk characteristics are derived off-line and stored in files/tables for access during run-time estimation. A time (RT) to rotate the disk from a reference sector to each physical sector on the hard drive is derived and stored for each logical block address (LBA). A time (HT) for the heads to move an incremental number of cylinders also is derived. Access time of a destination LBA after completing access to a source LBA is estimated using estimated rotational times for the source and destination looked up in an RT table, and a head movement time estimate looked up in an HT table, based on a run-time calculated cylinder increment between the source and destination LBAs.

Abstract: A cache for improving access to optical media includes a primary cache comprising RAM and a secondary cache comprising a portion of hard disk memory. Multiple aspects of the invention are defined: (1) Cache data discrimination: Discrimination methodology is implemented for determining when data should not be cached. Under certain conditions, caching of data is less likely to improve access time. (e.g., when the transfer rate is already exceeding a critical sustained throughput rate; when an estimated time to complete a CD-ROM data request is within a specific percentage of the estimated time to complete a hard drive disk request). (2) Secondary cache fragmentation avoidance: To keep the access time to secondary cache faster than the access time to the optical media, fragmentation of the secondary cache (i.e., hard disk) is minimized.

Abstract: A discrimination method and apparatus for determining when data should not be cached. Under certain conditions, caching of data is less likely to improve access time, (e.g., when the transfer rate is already exceeding a critical sustained throughput rate; when an estimated time to complete a CD-ROM data request is within a specific percentage of the estimated time to complete a hard drive disk request).

Abstract: A cache for improving access to optical media includes a primary cache comprising RAM and a secondary cache comprising a portion of hard disk memory. Multiple aspects of the invention are defined: (1) Cache data discrimination: Discrimination methodology is implemented for determining when data should not be cached. Under certain conditions, caching of data is less likely to improve access time. (e.g., when the transfer rate is already exceeding a critical sustained throughput rate; when an estimated time to complete a CD-ROM data request is within a specific percentage of the estimated time to complete a hard drive disk request). (2) Secondary cache fragmentation avoidance: To keep the access time to secondary cache faster than the access time to the optical media, fragmentation of the secondary cache (i.e., hard disk) is minimized.

Abstract: A cache for improving access to optical media includes a primary cache comprising RAM and a secondary cache comprising a portion of hard disk memory. Multiple aspects of the invention are defined: (1) Cache data discrimination: Discrimination methodology is implemented for determining when data should not be cached. Under certain conditions, caching of data is less likely to improve access time. (e.g., when the transfer rate is already exceeding a critical sustained throughput rate; when an estimated time to complete a CD-ROM data request is within a specific percentage of the estimated time to complete a hard drive disk request). (2) Secondary cache fragmentation avoidance: To keep the access time to secondary cache faster than the access time to the optical media, fragmentation of the secondary cache (i.e., hard disk) is minimized.