Abstract: An optimizer process in a storage system automatically selects access activity data for storage devices in the system during periods of interest so that a volume-swapping optimization analysis is based on desired device performance information and thus yields improved optimization results. For each of a number of sampling intervals in an analysis period, workload statistics are collected for logical volumes in the system. From the workload statistics, a service time is calculated for each storage device for each sampling interval. A set of highest service times is identified from the service times for each device, such as those above the 80th percentile, and a service time measure is generated from the set of highest service times.

Abstract: During sampling intervals, pairs of swap scores are assigned to respective pairs of storage devices of a storage system, each swap score pair indicating an amount of system performance improvement for a swap of logical volumes between source and target storage devices of the pair of storage devices. The swap scores are summed over all the intervals. A subset of the storage devices are then selected for a full optimization process based on the summed swap scores, where the full optimization process exhaustively looks for some number of best swaps among the storage devices to improve system performance. By choosing the size of the subset of storage devices, the processing burden required to perform the analysis will be in line with the processing capacity of the processing platform on which the analysis is performed, while achieving system performance improvement commensurate with a worst-case process in which every storage device in the system is analyzed for candidate swaps.

Abstract: In a selective logical-volume swapping process, a subset of storage devices in a storage system are selected that represent good candidates for swaps that will improve system performance. Workload statistics are utilized from a number of sample intervals in a relatively long analysis interval. The workload statistics are aggregated over intermediate intervals of each analysis interval to yield a set of aggregated statistics much smaller than the set of workload statistics. Based on the aggregated statistics, a service processor searches for swaps of logical volumes, ranks the swaps according to expected system performance improvement, and selects source and target storage devices of the higher-ranked swaps. The service processor can then perform a full optimization analysis for this subset of storage devices utilizing the workload statistics from all the sample intervals.

Abstract: In an optimization dispatch process for a storage system, M storage devices are identified for an optimization analysis process to be executed to identify logical volume swaps for improving system performance. The M storage devices are identified by merging candidate storage devices obtained from separate identification algorithms employing distinct identification criteria. The analysis process is dispatched for execution based on the M storage devices until (i) a successful completion upon which a best swap has been identified, or (ii) a resource-based stop condition such as an execution time limit. M is adjusted for a subsequent iteration such that over time a desired rate of successful completions is achieved. For example, M is increased by a first amount upon a successful completion and decreased by a second amount upon the stop condition.

Abstract: Optimizing performance of physical disk drives in a disk array storage device with a plurality of logical volumes is accomplished by accumulating statistics over an interval to select two logical volumes for an exchange. After testing to determine any adverse effect of making that change, the exchange occurs to more evenly distribute the loading on individual physical disk storage devices. One set of statistics, seek time, is obtained by combining estimated disk access information and an interpolated seek time obtained from a template that stores characteristics seek times among different pairs of segments that define a physical disk device independently of a logical volume configuration.

Abstract: A process that swaps physical storage volumes. The process performs a number of simple arithmetic operations to find a seek time for B new states of a disk. The number is of order B. Each new state is related to a current state of the disk by a swap with a selected external storage volume. The process also includes swapping one of the storage volumes of the disk with the selected storage volume in response to determining that the state produced by the swap has a smallest seek or access time among the new states.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. A list of exchangeable pairs of blocks is developed based on size and function. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. A statistical analysis leads to a selection of one block pair. After testing to determine any adverse effect of making that change, the exchange is made to more evenly distribute the loading on individual physical disk storage devices.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. A method is disclosed for efficiently generating disk access time based upon these statistics.

Abstract: Described are techniques for determining temporary storage areas for logical volumes to be swapped. Logical volumes may be swapped in a computer system in connection with efforts to improve performance of associated data storage entities in the computer system. As part of the data swapping and physical relocation of logical volumes, temporary storage locations are used. A temporary storage location is associated and used with each logical volume being swapped. Determining a temporary storage area for a logical volume may be made in accordance with particular criteria. A temporary storage area is determined for each logical volume from a list of one or more possible candidate locations. The problem of matching a logical volume to a unique temporary storage area may be represented as a bipartite graph matching problem using solutions, such as the maximum flow, and other techniques to find a match for each logical volume.

Abstract: Described are techniques used in connection with determining one or more pairs of logical volumes (LVs) to increase the performance of storage devices by swapping these LVs. Future performance of the anticipated potential LV swap is determined in accordance with past performance data collected prior to actually performing the swap. Data is gathered for a first arrangement of data storage devices for a period of time. A proposed modification to the first arrangement, such as swapping LVs of data on one or more storage devices, is produced. A performance estimation is produced using past performance data for the one or more pairs of proposed LV swap candidates. The search for pairs of LV candidates is determined using an iterative technique in which earlier iterations attempt to produce candidates from a smaller, more focused pool of possible LVs. The pool of LVs is expanded upon subsequent iterations if earlier iterations using heuristics providing for a more focused search are unsuccessful.

Abstract: A process performs an intra-disk swap. The process includes finding a set of values indicative of access loads of new states of a disk. Each new state is produced from a current state of the disk by hypothetically swapping a pair of physical storage volumes of the disk. The process also includes performing the swap of a pair of physical storage volumes based on the value of the access load of the new state produced by hypothetically swapping the pair.

Abstract: A process determines a seek time of an access head of a physical storage disk. The process includes collecting statistical access data on physical storage volumes served by the access head and determining seek times for a set of reference head travel time functions. Each function depends on distances between pairs of the physical storage volumes. The process also includes performing a sum of the determined seek times. Each term of the sum is weighted by an expansion coefficient of a travel time function of the access head with respect to the set of reference head travel time functions.

Abstract: A mechanism for adjusting seek activity in a data storage system of physical devices having mirrored logical volumes is presented. Statistics describing at least reading data from the mirrored volumes during successive time periods are collected. From the collected statistics an activity level associated with each of the mirrored logical volumes is determined. Seek activity values for the physical devices are computed based on the activity levels associated with the logical volumes stored on each of the physical devices. The computed seek activity values relate a physical device seek activity to the activity level associated with, and distance between, the mirrored logical volumes residing the physical devices. The computed seek values are used to minimize seek activity for non-mirrored ones of the physical devices.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. A list of exchangeable pairs of blocks is developed based on size and function. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. These activities are represented as disk seek, latency and data transfer times. A statistical analysis leads to a selection of one block pair. After testing to determine any adverse effect of making that change, the exchange is made to more evenly distribute the loading on individual physical disk storage devices.

Abstract: A method checks a swap's quality. The method includes finding an upper bound for reductions to a disk's seek time or total access time. The reductions are caused by swapping single storage volumes of the disk, which is in a preselected state, with an external storage volume. The method also includes finding a reduction to the seek time or total access time produced by swapping one storage volume of the disk with a selected external storage volume and comparing the found reduction to the upper bound.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. A method is disclosed for efficiently generating disk access time based upon these statistics.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. A list of exchangeable pairs of blocks is developed based on size and function. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. These activities are represented as disk seek, latency and data transfer times. A statistical analysis leads to a selection of one block pair. After testing to determine any adverse effect of making that change, the exchange is made to more evenly distribute the loading on individual physical disk storage devices.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. A list of exchangeable pairs of blocks is developed based on size and function. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. These activities are represented as disk seek, latency and data transfer times. A statistical analysis leads to a selection of one block pair. After testing to determine any adverse effect of making that change, the exchange is made to more evenly distribute the loading on individual physical disk storage devices.

Abstract: A method and an apparatus for reconfiguring a storage subsystem by performing an ordered sequence of reconfigurations of physical storage volumes of the storage subsystem. The method and apparatus perform a portion of the sequence of reconfigurations, in response to receiving a rollback request, in an order that is reversed with respect to the order of the sequence.

Abstract: Load balancing of activities on physical disk storage devices is accomplished by monitoring reading and writing operations to blocks of contiguous storage locations on the physical disk storage devices. Statistics accumulated over an interval are then used to obtain access activity values for each block and each physical disk drive. A method is disclosed for efficiently generating disk access time based upon these statistics.