Patent Issued for System and Method for Selecting and Scheduling Corrective Actions for Automated Storage Management

By a News Reporter-Staff News Editor at Information Technology Newsweekly -- According to news reporting originating from Alexandria, Virginia, by VerticalNews journalists, a patent by the inventors Korupolu, Madhukar R. (Sunnyvale, CA); Uttamchandani, Sandeep M. (San Jose, CA); Voruganti, Kaladhar (San Jose, CA); Yin, Li (Redmond, WA), filed on April 19, 2007, was cleared and issued on December 4, 2012.

The assignee for this patent, patent number 8326669, is International Business Machines Corporation (Armonk, NY).

Reporters obtained the following quote from the background information supplied by the inventors: "The embodiments of the invention generally relate to storage system management, and, more particularly, to a system and a method for selecting and scheduling corrective actions to solve current and anticipated service level objective violations in a data storage system.

"Currently, there are different types of planning tools that can be used by storage system administrators to develop corrective action plans in response to identified current and anticipated workload service level objective (SLO) violations. These planning tools include, but are not limited to, workload throttling planning tools, buffer size planning tools, data migration planning tools and new hardware planning tools. A major limitation with these planning tools is that the storage system administrator must manually decide which tools to select and when to deploy the corrective action plans developed by those tools in order to solve an identified workload SLO violation. Manual selection of a planning tool and of corrective action plans is increasingly becoming difficult due to an increase in the number of storage resources, applications, and business policies."

In addition to obtaining background information on this patent, VerticalNews editors also obtained the inventors' summary information for this patent: "In view of the foregoing, disclosed herein is a storage management framework that integrates corrective action plans output from multiple different types of planning tools, sorts the different corrective action plans based on utility and risk values and provides a time-based schedule for implementing one or more of the corrective action plans.

"Embodiments of the management system can comprise a controller, multiple different corrective action planning tools (e.g., a workload throttling planning tool, a buffer size planning tool, a data migration planning tool, a new hardware planning tool, etc.) and a translator for providing communication between the controller and the different planning tools.

"The controller can receive input from the storage system administrator as well as directly from the data storage system. This input can include, but is not limited to, administrator-defined business-level constraint information, system state information and triggering event information. The administrator-defined business-level constraints can comprise per-workload utility functions, budget constraints, a desired analysis window and service-level objectives (SLOs) information (e.g., workload SLOs and violation thresholds for those workload SLOs). System attributes can comprise the current system state and the time-series values of workload request-rates (i.e., anticipated states in the future). Specifically, these states can comprise the hardware components (arrays, switches etc.) and their utilization; the current and anticipated workload characteristics (e.g., IOPS, read write ratio, request size, random sequential ratio); and a mapping of workloads to components. Triggering event details can comprise the triggering event type (e.g., proactive in response to an anticipated SLO violation or reactive in response to a current SLO violation) and the possible affected components.

"The controller can further evaluate the inputs and, particularly, evaluate the triggering event details in light of the business constraints to determine if corrective action is warranted (e.g., based on actual or anticipated system utility loss as a result of the current or anticipated workload SLO violations, respectively).

"The controller can also generate a query and communicate that query to the different planning tools via the translator, if corrective action is deemed warranted. Specifically, the query can comprise a selection of planning tools to which the query is directed, the current and anticipated states of the system and the utility functions of the workloads. Additionally, rather than a single query, the controller can generate a plurality of queries to be directed to each of the planning tools. These queries can each specify a different time period and request a corrective action plan that is optimized for that specified time period.

"Each of the planning tools can develop corrective action plans that can be implemented in response to the triggering event and, if applicable, that are optimized for a time period specified in the query.

"The translator allows communication between the controller and the various planning tools. That is, both the controller and the planning tools can be in communication with the translator. The controller can forward the query or queries to the translator, where they are translated into the appropriate formats for input into the different planning tools. As the corrective action plans are developed by the planning tools, they can be forwarded to the risk-utility calculator.

"Each corrective action plan can comprise a recommendation for a corrective action along with the parameters of the action, approximate times for initiating and completing the action, costs for implementing the corrective action, predicted performance of workloads in response to the action and, if applicable, the specified time period for which the corrective action plan was generated.

"The risk-utility calculator calculates risk and utility values for each of the corrective action plans recommended by each of the planning tools, based on the business constraints and on the system attributes. The utility value of a corrective action plan can comprise an expected benefit to the data storage system if the corrective action plan is implemented and the risk value can comprise the probability that the expected benefit will be lost as a result of volatility in workload time-series functions and/or as a result of human error. Once the utility and risk values are determined, the risk-utility calculator can forward a list of the corrective action plans, along with their corresponding utility and risk values, to the controller.

"The controller can then create and output a schedule for implementing at least one of the corrective action plans recommended based on the risk and utility values. Specifically, the controller can create and output a time-based schedule specifying one or more corrective action plans, specifying the sequence for the implementing those corrective action plans(s) and specifying the approximate times for initiating and completing performance of those corrective action plan(s). This schedule can also set out checkpoint states (i.e., a set of anticipated states that the data storage system will exhibit as the corrective action plans are implemented according to the schedule).

"The management system can also comprise at least one monitor for continuously monitoring the state of the data storage system and forwarding this state information to the controller. Thus, as corrective action plans are implemented according to the schedule, the controller can determine the impact on the data storage system of corrective action implementation and can further alter the schedule, as necessary (e.g., if the actual state of the data storage system at a given time differs significantly from the checkpoint state for that given time).

"Embodiments of the method of managing a system as described above can comprise receiving inputs from the storage system administrator as well as directly from the data storage system itself. These inputs can comprise administrator-defined business-level constraints, system attributes and a triggering event, as described above.

"The method can further comprise automatically evaluating the inputs and, particularly, evaluating the triggering event details in light of the business constraints to determine whether or not corrective action is warranted (e.g., based on actual or anticipated system utility loss as a result of the current or anticipated workload SLO violations, respectively). If corrective action is warranted, the method can comprise automatically generating a query. This query can comprise a selection of the planning tools to which the query should be directed, the current and anticipated states of the data storage system, and the utility functions of the workloads. Additionally, rather than a single query, a plurality of queries can automatically be generated, each identifying a different time period (e.g., a day, a week, a month, a year and/or any other specified number of days) and requesting a corrective action plan that is optimized for that time period.

"Once generated, the query or queries can be automatically translated into the appropriate formats for input into each of the specified planning tools and can be communicated to those different planning tools. Corrective action plans are then developed by and received back from each of the planning tools in response to the query/queries.

"Each of the corrective action plans can comprise a recommendation for a corrective action along with the parameters of the action, approximate times for initiating and completing the action, costs, and predicted performance of workloads in response to the action. If applicable, each plan can also indicate the specified time period for which it was optimized.

"Upon receipt of the corrective action plans, risk and utility values are automatically calculated for each of the recommended corrective action plans from each of the different planning tools based on the parameters of the actions, the business constraints and the system attributes.

"Once the utility and risk values are determined, a schedule for performing at least one of these corrective actions can be created and output based on the risk and utility values. Creating the schedule comprises selecting one or more of the corrective action plans, determining the sequence for implementing those corrective action plans(s), and determining the approximate times for initiating and completing performance of those corrective action plan(s). Checkpoint states (i.e., a set of anticipated states that the data storage system will exhibit as the corrective action plans are implemented according to the schedule) can also be determined.

"Implementation of the corrective action plans according to the schedule can be automated and/or manual.

"The method can further comprise continuously monitoring the state of the data storage system and determining the impact on the data storage system of implementing the corrective action plans. Thus, the schedule can be automatically altered, as necessary (e.g., if the actual state of the data storage system at a given time differs significantly from the checkpoint state for that given time).

"Finally, also disclosed are embodiments of a program storage device readable by computer and tangibly embodying a program of instructions executable by the computer to perform the method described above.

"These and other aspects of the embodiments of the invention will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating preferred embodiments of the invention and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments of the invention without departing from the spirit thereof, and the embodiments of the invention include all such modifications."

For more information, see this patent: Korupolu, Madhukar R.; Uttamchandani, Sandeep M.; Voruganti, Kaladhar; Yin, Li. System and Method for Selecting and Scheduling Corrective Actions for Automated Storage Management. U.S. Patent Number 8326669, filed April 19, 2007, and issued December 4, 2012. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=25&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=1216&f=G&l=50&co1=AND&d=PTXT&s1=20121204.PD.&OS=ISD/20121204&RS=ISD/20121204

Keywords for this news article include: Information Technology, Information and Data Storage, Information and Data Migration, International Business Machines Corporation.

Our reports deliver fact-based news of research and discoveries from around the world. Copyright 2012, NewsRx LLC