Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS20070130236 A1
Publication typeApplication
Application numberUS 11/294,165
Publication dateJun 7, 2007
Filing dateDec 5, 2005
Priority dateDec 5, 2005
Also published asCN1979483A, US20110029419
Publication number11294165, 294165, US 2007/0130236 A1, US 2007/130236 A1, US 20070130236 A1, US 20070130236A1, US 2007130236 A1, US 2007130236A1, US-A1-20070130236, US-A1-2007130236, US2007/0130236A1, US2007/130236A1, US20070130236 A1, US20070130236A1, US2007130236 A1, US2007130236A1
InventorsJames Seeger, Jason Young
Original AssigneeInternational Buisiness Machines Corporation
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Method, apparatus and program storage device for providing real-time file system charge-back accounting per management object during a report cycle
US 20070130236 A1
Abstract
A method, apparatus and program storage device for providing real-time file system charge-back accounting with real time historical minimum and maximum usage per user or group during a report cycle. In a first storage area of file system memory, space allocation data for a current accounting period is monitored and persistently maintained. A minimum, maximum and average space allocation is calculated for the management object during the current accounting period. The calculated minimum, maximum and average space allocation for the management object may be moved to a second storage area of the file system memory while beginning to monitor and persistently maintain space allocation data of the management object in the first storage area of the file system memory during a subsequent accounting period.
Images(10)
Previous page
Next page
Claims(21)
1. A file system for providing charge back accounting data, comprising a first file system buffer for persistently maintaining space allocation data during a current accounting period, wherein the space allocation data comprises a minimum, a maximum and an average space allocation for a management object during the current accounting period.
2. The file system of claim 1, wherein the space allocation is transactionally and persistently maintained per management object.
3. The file system of claim 2, wherein the space allocation per management object is maintained per storage tier.
4. The file system of claim 1, wherein the average space allocation comprises an average that is weighted based on lengths of time and size of space allocation during the current accounting period.
5. The file system of claim 1, wherein space allocation data in the first file system buffer is updated as space is allocated and deallocated.
6. The file system of claim 1 further comprising a second file system buffer for persistently maintaining space allocation data for a previous accounting period; wherein the space allocation data persistently maintained during the first accounting period is moved to the second file system buffer at the end of the current accounting period.
7. A storage system, comprising:
a pool of storage;
a controller, coupled to the pool of storage, for controlling user access to the storage pool;
a file system for providing charge back accounting data based upon space allocation; and
an accounting system for using the file system charge back accounting data to generate a report describing storage space allocation by a management object;
wherein the file system comprises a first file system buffer for persistently maintaining space allocation data during a current accounting period, the space allocation data comprising a minimum, a maximum and an average space allocation for a management object during the current accounting period.
8. The storage system of claim 7, wherein the space allocation is transactionally and persistently maintained per management object.
9. The storage system of claim 8, wherein the space allocation per management object is maintained per storage tier.
10. The storage system of claim 7, wherein the average space allocation comprises an average that is weighted based on lengths of time and size of space allocation during the current accounting period.
11. The storage system of claim 7, wherein space allocation data in the first file system buffer is updated as space is allocated and deallocated.
12. The storage system of claim 7, wherein accounting system applies cost constants to the space allocation data to capture storage attributes.
13. The storage system of claim 12, wherein the storage attributes comprise at least one selected from the group comprising a quality of service, service level agreements, and tiered storage class.
14. The storage system of claim 7 further comprising a second file system buffer for persistently maintaining space allocation data for a previous accounting period, wherein the space allocation data persistently maintained during the first accounting period is moved to the second file system buffer at the end of the current accounting period.
15. A program storage device, comprising:
program instructions executable by a processing device to perform operations for providing file system charge back accounting data, the operations comprising:
monitoring and persistently maintaining, in a first storage area of file system memory, space allocation data for a current accounting period;
calculating a minimum, maximum and average space allocation for a management object during the current accounting period; and
generating a report based upon the calculated minimum, maximum and average space allocation for the management object during the current accounting period.
16. The program storage device of claim 15, wherein the calculating a minimum, maximum and average space allocation for the management object during the ended current accounting period further comprises calculating an average that is weighted based on lengths of time and size of space allocation during the current accounting period.
17. The program storage device of claim 15, wherein the monitoring and persistently maintaining, in a first storage area of file system memory, space allocation data for a current accounting period further comprises transactionally and persistently maintaining space allocation data per management object.
18. The program storage device of claim 15, wherein the monitoring and persistently maintaining, in a first storage area of file system memory, space allocation data for a current accounting period further comprises updating the space allocation data as space is allocated and deallocated.
19. The program storage device of claim 15 further comprising:
persistently maintaining space allocation data for a previous accounting period in a second storage area of the file system memory; and
moving the maintained space allocation for the management object to the second storage area of the file system memory while beginning to monitor and persistently maintain space allocation data of the management object in the first storage area of the file system memory during a subsequent accounting period.
20. A method for providing file system charge back accounting data, comprising:
monitoring and persistently maintaining, in a first storage area of file system memory, space allocation data for a current accounting period;
calculating a minimum, maximum and average space allocation for a management object during the current accounting period; and
generating a report based upon the calculated minimum, maximum and average space allocation for the management object during the current accounting period.
21. A method for providing file system charge back accounting, comprising:
receiving metadata representing space allocation per management object based on file system activity;
persistently maintaining in a storage area the metadata representing the space allocation for a current accounting period; and
using the received metadata to calculate a minimum, maximum and average space allocation for the management object during the current accounting period.
Description
    BACKGROUND OF THE INVENTION
  • [0001]
    1. Field of the Invention
  • [0002]
    This invention relates in general to a storage system accounting, and more particularly to a method, apparatus and program storage device for providing real-time file system charge-back accounting with real time historical minimum and maximum usage per user or group during a report cycle.
  • [0003]
    2. Description of Related Art
  • [0004]
    In today's information age, computer and communication systems often require large amounts of data storage. More and more frequently, access to data storage is being provided over communication networks. Among other things, this makes data storage resources available to a number of data storage users.
  • [0005]
    Storage infrastructures of enterprises are feeling a strain as the huge demand for storage continues to rapidly increase. Escalating capacity and bandwidth requirements cause costs to soar as resource-constrained organizations struggle to manage the complexity and sheer volume of data that must be stored, accessed, and protected.
  • [0006]
    The increase in average file size (text-based to multimedia) continues to grow. New applications are driving storage capacity upward and introducing new types of data. At the same time, Internet access devices, such as mobile phones and Personal Digital Assistants (PDAs), are getting smaller with less processing power, less storage, and smaller screen sizes. This trend implies that more content will have to be processed and stored rather than within the access device itself.
  • [0007]
    Most businesses today are faced with high growth in the amount of data they have to store, but IT budgets are flat, or being reduced. IT can no longer afford to keep all their data on high-end storage platforms or staffing levels to manage large numbers of storage targets. The answer is to produce a small set of storage tiers that align with business requirements to provide the optimum service level at the lowest cost. Tiered Storage School will help in understanding which technologies are available for implementing a tiered storage strategy.
  • [0008]
    Information Lifecycle Management (ILM) is a process that includes Information Technology (IT) staff making decisions about what data should reside where in the physical infrastructure. ILM is a process in which people, processes, and technologies deal with data from cradle to grave. Data has different value at inception than it does at death, so it shouldn't be treated the same way during each phase of its lifetime.
  • [0009]
    However, a distinction may be made between Data Lifecycle Management (DLM) and ILM. DLM is the infrastructural element and ILM is the macro-level element that assigns both subjective and objective value. ILM relates not only to an application, but also to where the information itself resides in the company as a whole. DLM, which involves the infrastructural component of what goes where and when, is very similar to the traditional Hierarchical Storage Management (HSM) that has existed in the mainframe world since inception. The primary difference is that HSM is a single-server, single storage stack methodology for moving data down the physical stack in order to lower the cost of primary storage, whereas DLM architectures automate the processes involved, typically organizing data into separate tiers according to specified policies, and automating data migration from one tier to another based on those criteria. DLM includes two tiers. There is the physical tier of actual storage where data resides, including disk arrays of all sizes and tapes, and then there is the process tier for moving data between those elements.
  • [0010]
    Companies today are constructing tiered storage infrastructures. These companies continue to deploy storage systems that deliver many different classes of storage ranging from high performance/high cost to high capacity/low cost. Through the deployment of SANs, many of these storage assets are now physically connected to servers that run many different types of applications and create many kinds of information. Finally, with the arrival of network-resident storage services for distributed management of volume, files, and data replication managers can intelligently manage storage assets to meet the needs of many different applications across the network instead of device by device.
  • [0011]
    Accordingly, the growing importance of data as a corporate asset puts increasing pressure on organizations to protect and manage storage to ensure its continuous availability. As sites are faced with deploying larger and more complex storage infrastructure, storage management costs compound quickly and technology planning becomes riskier. Moreover, the rapid evolution of technology mandates training of specialized personnel to maximize performance gains and ensure compatibility with emerging standards.
  • [0012]
    Because of these factors, it is important that today's accounting process measure and assign resource and service utilization accurately. An effective way to measure, report, and charge for the operational costs and services is to integrate the accounting process with the measurement system. As a result, the role of accounting has expanded from capacity planning, resource utilization, and performance monitoring to one of measuring and charging operational costs for technical resources and end user services provided by data center. Accounting process must gather accounting information from different sources. The accounting process needs to be closely linked with both the Service Level Agreements (SLA) and the reporting process to successfully manage and charge storage related costs.
  • [0013]
    With data processing, each department and every project produces costs. The goal of the accounting and chargeback system is to identify total costs for each cost categories, what caused the costs, who caused the costs, how to distribute the cost and who to bill. An essential requirement of the accounting and chargeback system is to accurately identify all costs, measure the usage and correctly distribute the costs. All resources and services must be analyzed to determine who is consuming the resource or service and what costs are associated with the resource or service so the cost can be charged. This process is not simple, and requires detailed knowledge of the systems to recognize the correct data and correctly associate the data in the accounting and chargeback system.
  • [0014]
    The main purpose of an accounting system is chargeback. Chargeback is when the cost of services provided are identified and charged to the users and recipients of the services. Without charging for the services, the process of accounting for services is of little value. The accounting system is used to discover resource usage and assign the various resource usages to different causes. By assigning a value to each resource usage, a chargeback system is developed.
  • [0015]
    File system chargeback accounting is point-in-time in nature with today's file systems. Normally the functionality is tied to a file system's user or group quota and some independent reporting utility. The space reported is generated from the current quota accounting within the file system via some customized scripting. The major drawback is space utilization can dynamically change, sometimes significantly, during a reporting cycle. There is no way to determine that a user has potentially large cyclic space usage. Additionally, some quota implementations only report the used disk space and do not consider allocated space.
  • [0016]
    Another solution that exists in some space reporting tools today is a sampling method which estimates the amount of space used by a user or application. The problem with this approach is the sample period can be very costly and time consuming. Additionally, the results are only an estimation of the space utilization. Additionally, this type of method can only categorize file systems and applications it knows about. New file systems or application support in a customer environment would need to be added. Some SAN providers provide chargeback at a logical LUN granularity, but not at a file system user or group granularity.
  • [0017]
    It can be seen then that there is a need for a method, apparatus and program storage device for providing file system chargeback accounting with real time historical minimum and maximum usage per management object, e.g., user or group, during a report cycle.
  • SUMMARY OF THE INVENTION
  • [0018]
    To overcome the limitations in the prior art described above, and to overcome other limitations that will become apparent upon reading and understanding the present specification, the present invention discloses a method, apparatus and program storage device for providing real-time file system charge-back accounting with real time historical minimum and maximum usage per management object during a report cycle.
  • [0019]
    The present invention solves the above-described problems by providing an arrangement for storing space allocation data for users in file system memory. This allows an administrator to define a charge-back accounting cycle. Once a new accounting cycle is defined, the minimum and maximum space utilized by each user is maintained for reporting until the next accounting cycle. An average space utilization value may be calculated, e.g., mean, weighted average, etc, to allow generation of charge-back accounting. The charge-back data is maintained real time with file system activity to provide a more accurate space utilization report. The space allocation data is persistently maintained during a first accounting period in a first file system buffer and then moved to a second file system buffer at the end of the current accounting period. This move may be physical or logical.
  • [0020]
    A file system for providing charge back accounting data in accordance with the principles of the present invention includes a first file system buffer for persistently maintaining space allocation data during a current accounting period, wherein the space allocation data comprises a minimum, a maximum and an average space allocation for a management object during the current accounting period.
  • [0021]
    In an embodiment of the present invention implemented as a storage system, an accounting system is provided that uses the file system charge back accounting data to generate a report describing storage space allocation by a management object, e.g., a user or a group. The file system for providing charge back accounting data includes a first file system buffer for persistently maintaining space allocation data during a current accounting period, the space allocation data comprising a minimum, a maximum and an average space allocation for a management object during the current accounting period. Embodiments implementing a method, a service and a program storage device that includes program instructions executable by a processing device to perform operations for providing file system charge back accounting data are also described.
  • [0022]
    These and various other advantages and features of novelty which characterize the invention are pointed out with particularity in the claims annexed hereto and form a part hereof. However, for a better understanding of the invention, its advantages, and the objects obtained by its use, reference should be made to the drawings which form a further part hereof, and to accompanying descriptive matter, in which there are illustrated and described specific examples of an apparatus in accordance with the invention.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • [0023]
    Referring now to the drawings in which like reference numbers represent corresponding parts throughout:
  • [0024]
    FIG. 1 illustrates a block diagram of a hierarchical data storage management system according to one embodiment of the present invention;
  • [0025]
    FIG. 2 represents a data storage hierarchy according to an embodiment of the present invention;
  • [0026]
    FIG. 3 illustrates a three-tiered storage model according to one embodiment of the present invention;
  • [0027]
    FIG. 4 illustrates a SAN that may be configured to provide a file system that provides file system charge-back accounting according to an embodiment of the present invention;
  • [0028]
    FIG. 5 illustrates a file system charge back buffer arrangement for providing a record of space utilization as part of a file system's metadata information according to an embodiment of the present invention;
  • [0029]
    FIG. 6 is a flow chart showing the method for calculating a weighted average of space allocation according to an embodiment of the present invention;
  • [0030]
    FIG. 7 is a flow chart of the method for providing file system chargeback accounting according to an embodiment of the present invention; and
  • [0031]
    FIG. 8 illustrates a system according to an embodiment of the present invention.
  • DETAILED DESCRIPTION OF THE INVENTION
  • [0032]
    In the following description of the embodiments, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration the specific embodiments in which the invention may be practiced. It is to be understood that other embodiments may be utilized because structural changes may be made without departing from the scope of the present invention.
  • [0033]
    The present invention provides a method, apparatus and program storage device for providing real-time file system charge-back accounting with real time historical minimum and maximum usage per management object, e.g., user or group, during a report cycle. An arrangement is provided for storing space allocation data for users in file system memory. This allows an administrator to define a charge-back accounting cycle. Once a new accounting cycle is defined, the minimum and maximum space utilized by each user is maintained for reporting until the next accounting cycle. An average space utilization value may be calculated, e.g., mean, weighted average, etc, to allow generation of charge-back accounting. The charge-back data is maintained real time with file system activity to provide a more accurate space utilization report.
  • [0034]
    FIG. 1 illustrates a block diagram of a hierarchical data storage management system 100 according to one embodiment of the present invention. The system 100 generally includes a hierarchical data storage subsystem 104 that is accessible by clients 102. The client stations 102 may be local or remote relative to the data processing apparatus 120 and generally access the data processing apparatus 120 via a network 110. The hierarchical data storage subsystem 104 is configured to store data and manage the stored data according to storage access requests from user client stations 102. The depicted storage subsystem 104 includes a data processing apparatus 120 operationally coupled to one or more hierarchical data storage units 122 via a communications channel 126. The communications channel 126 may be a storage area network (SAN) or alternately may be another type of communications channel.
  • [0035]
    The data processing apparatus 120 includes a central processing unit 130 for processing the digital signals received from client stations 102. The central processing unit 130 is coupled with an I/O processor 132. The central processing unit 130 transmits signals to the hierarchical data storage units 122 and database 124 via the I/O processor 132 and communications channel 126. The central processing unit 130 is also digitally coupled to memory storage 140 for storing programming instructions 142 accessed by the central processing unit 130 for control of the digital processing system 120.
  • [0036]
    The data processing apparatus 120 also includes a storage management system 160 and an accounting system 180. The management system 160 monitors usage of all devices and outputs information indicative of such use in a common format. In general, the management system can provide data for a variety of different applications, aside from the billing application illustrated here. The accounting system 180 creates and stores usage information. Accordingly, the accounting system 180 bills each customer based on a calculation as will be described herein below. Accounting system 180 executes an accounting process that collects performance status to derive resource utilization of the hierarchical data storage units 122. The collected performance status is maintained persistently in a dual buffer system as described herein below.
  • [0037]
    FIG. 2 represents a data storage hierarchy 200 according to an embodiment of the present invention. In FIG. 2, diagrammatically “higher” data storage media and devices correspond to faster accessibility to stored data, disaster recovery and nearline restore capabilities. Specifically, this depiction includes high-speed data storage media and devices at the “top” levels 202 and 204 of the hierarchy 200. For example, level 202 might include a direct access storage device (DASD) such as a high-speed magnetic disk drive or high-speed optical disks and drives. Level 204 might include storage media and devices similar to those in level 202, but of slower access speeds.
  • [0038]
    In FIG. 2, level 206 includes multiple optical disks and one or more corresponding optical disk drives. Once again, these storage media devices represent access times slower than the devices depicted in levels 202 and 204. Levels 208 and 210 represent the slowest access times for all of the media types and devices shown in the depicted storage hierarchy 200. These levels 208 and 210 might include sequential access storage devices such as magnetic tape media and drives.
  • [0039]
    The storage hierarchy 200 illustrated in FIG. 2 is also very helpful to illustrate one example of a cost structure for various media types and devices within the storage hierarchy 200. For example, the “bottom” levels 210 and 208 of the diagram may represent the least costly storage implementation per data unit while the “top” levels 202 and 204 may represent the most costly data storage schemes. From this it is apparent and not unexpected that the storage media devices that offer the fastest data access times are also typically the most expensive to implement for a given amount of data storage capacity.
  • [0040]
    Thus, a client with very stringent performance or recovery requirements in need of the absolutely fastest data retrieval available would not be satisfied with the implementation of a system consisting solely of currently available magnetic tapes and drives. Instead, such a client would desire access to direct access storage devices or other more costly storage resources for all data storage within the projected data storage capacity requirements and project funding constraints. Moreover, a user on a client may have files that exist in one, some or all of the storage tiers.
  • [0041]
    Another aspect of the storage hierarchy 200 is the designation and use of storage pools. A storage pool is one or more storage media, such as disks and tapes, that are assigned as a group by the hierarchical storage manager for storage of similar data. The assignment may be automatically executed based on storage policy or may be manually dictated by a user via an administrator station. A typical storage pool may correspond to a particular type of data as illustrated in FIG. 3.
  • [0042]
    FIG. 3 illustrates a three-tiered storage model 300 according to one embodiment of the present invention. Data that is stored in a tiered or hierarchical storage system may be classified according to storage classes, which may be hierarchical attributes provided for each storage region based on the usage of data storage. For example, a three-tiered storage system 300 may be utilized, wherein the three attributes of online storage 310, nearline storage 320 and archive storage 330 are used. In addition, sub-attributes, such as premium 312 and normal 314 storage may be defined for the online storage.
  • [0043]
    Online storage 310 may be an attribute set for logical units (LUs) suitable for storing data of files that are frequently accessed, such as files being accessed online and files being generated. The premium sub-attribute 312 may indicate an attribute set for LUs suitable for storing data especially requiring fast response or more desirable recovery capabilities. Nearline storage 320 may be an attribute set for LUs suitable for storing data of files that are not frequently used but are occasionally accessed. Archive storage 330 may be an attribute set for LUs suitable for storing data of files that are hardly ever accessed and are maintained for long-term storage.
  • [0044]
    As can be seen in FIG. 3, the online 310 and nearline storage 320 may be implemented using network storage 340 and/or direct access storage 350. The network storage may further be implemented, for example, using disks 342 for the online storage 310 and disk and/or optical storage devices 344 for the nearline storage 320. Similarly, the direct access storage may further be implemented, for example, using disks 352 for the online storage 310 and disk and/or optical storage devices 354 for the nearline storage 320. As shown in FIG. 3, the offline storage 330 may be implemented using tape or other inexpensive storage devices 360.
  • [0045]
    While FIG. 3 shows a three-tiered storage model, those skilled in the art will recognize that embodiments of the present invention are not meant to be limited to a three-tiered model and that other models may be utilized in a manner that is consistent with the embodiments of the present invention described herein.
  • [0046]
    FIG. 4 illustrates a system 400 that may be configured to provide file system charge-back accounting according to an embodiment of the present invention. In FIG. 4, application, or client, servers 410-416, which request a file, obtain information about the file (the meta-data) from the servers 420 of the server cluster 402 that manages file locks and all other file information. The servers 420 then provide that information to the application, or client, servers 410-416. The application, or client, servers 410-416 then access the blocks comprising that file at the storage pools of the storage pools 430 directly through the SAN fabric 440.
  • [0047]
    A server 420 is designed to be implemented on a variety of operating systems, e.g., Windows to various flavors of Linux and UNIX, thereby allowing a server 420 and all of these operating systems to share files. In certain embodiments, server 420 may include a cluster of servers. For example, a file created in Windows will be as accessible from a Windows client, or application, server 416 as it is from Solaris 412, AIX 410, or any other supported platform, and vice versa. However, an application is still required to be able to read that file, no matter how accessible it is.
  • [0048]
    Since server 420 will have a complete understanding of all files managed by the server cluster 402, including the essential meta-data related to each file to make important decisions, a server 420 may be used to manage the storage on the network. A server 420 in server cluster 402 may be configured to provide space allocation control, policy management, read/write lock management, file directory and file system charge back records. Setting these policies relieves the storage administrator of the burden of repetitive tasks, and forms the basis of automation. Moreover, file system charge back accounting provided by embodiments of the present invention may manage user equivalence between operating systems.
  • [0049]
    A server 420 also provides the ability to group storage devices according to their characteristics, such as latency and throughput. These groupings, called storage pools 430, allow administrators to manage data according to predetermined characteristics. Because the meta-data for the server cluster 402 is separate from the application data, files can be manipulated while remaining active. When files are removed from service, the server cluster 402 may automatically reallocate the space without disruption. If a LUN is removed from the control of the server cluster 402, the data on that LUN can be automatically moved.
  • [0050]
    The server cluster 402 persistently maintains a record of space utilization as part of the file system's metadata information as users perform creation, allocation, deletion and ownership modification operations. A minimum and maximum space utilization is tracked per management object (user or group) for each reporting cycle. Optionally, time weighted averages can be generated based on real consumption over the reporting cycle. The charge-back feature can be integrated with other SAN configuration metadata to provide a resource aware pricing structure. For example, this could be accomplished by defining attributes of tiered storage when configuring pools and LUNs.
  • [0051]
    While the server 420 for controlling the metadata of the storage cluster 402 is shown in FIG. 4 as a separate entity, embodiments of the present invention are not meant to be limited to the use of a separate entity. Rather, the control provided by servers 420 may reside on any client, application, servers 410-416 in the network or even within a storage servers. Moreover, embodiments of the present invention are not meant to be limited to the storage architecture or hardware configuration illustrated in FIG. 4.
  • [0052]
    FIG. 5 illustrates a file system charge back buffer arrangement 500 for providing a record of space utilization as part of a file system's metadata information according to an embodiment of the present invention. While a double buffer is illustrated in FIG. 5, those skilled in the art will recognize that the present invention is not meant to be limited to a double buffer arrangement. For example, multiple buffers may be used to provide records over a plurality of accounting periods or the buffers may be subdivided into memory areas for maintaining space utilization records over a plurality of accounting periods. Alternatively, a single buffer may be used. A single buffer would still allow real-time minimum, maximum, and average space allocation data to be maintained, but would not be able to report on it statically.
  • [0053]
    In FIG. 5, the double buffer configuration 500 includes a first buffer 510 and a second 540 buffer for transactionally and persistently maintaining the space allocation records. The first buffer 510 persistently maintains a record of space utilization data for the current accounting period. The current copy buffer 510 maintains the record that gets updated as the user allocates and deallocates space. Thus, the first buffer 510 is dynamic. The double buffer 500 for implementing the system charge-back function keeps track of the minimum 520, maximum 522, and average 524 bytes used per user during a customer defined accounting period.
  • [0054]
    The second buffer 540 persistently maintains a record of space utilization data for a previous accounting period. Because the space utilization data of the second buffer 540 is for a previous accounting period, the second buffer 540 is static. Accordingly, the double buffer system 500 stores two copies of the minimum space utilization, maximum space utilization, and average bytes used values. Further, once an accounting period ends, the record in the current buffer 510 is moved to a buffer for maintaining a previous accounting period. This move may be physical or logical. When only two buffers are used, the data in the previous accounting period buffer 540 may be overwritten with the data moved form the current buffer 510 when the move is physical. However, a new buffer may be configured for maintaining the data being moved. Still further, a new buffer may be configured for the current accounting period and the old current buffer modified to be the most recent previous accounting period buffer.
  • [0055]
    Using the space utilization data 542 for a user that is maintained in the second buffer 540, an administrator may generate accounting reports on space usage based on the stable values in the previous accounting period. The same report can be generated several times within the same accounting period and provide the exact same results since it isn't subject to the most recent changes in space usage.
  • [0056]
    Accordingly, a file system charge-back configuration 500 according to an embodiment of the present invention allows the administrator to define a charge-back accounting cycle and independent report generation. A global state 570 which includes an index and the start times for the current and previous accounting periods is maintained. At the start of a new accounting period the index for the global state 570 is incremented. The start time is saved. Thus, the system stores the start time of the current period and the previous period.
  • [0057]
    Once a new accounting cycle is defined, the minimum and maximum space utilized by each user is maintained for reporting until the next accounting cycle. The file system may maintain the current space utilization information in the current accounting period buffer 510 along with a number of other records of previous space utilization 540. Additionally, the charge-back data of the current copy buffer is maintained real time with file system activity regarding the time 514 and space allocation 516 to provide a more accurate space utilization report. Furthermore, cost constants can be applied to actual storage allocation to capture specific storage attributes such as quality of service, service level agreements, or tiered storage class.
  • [0058]
    In FIG. 5, a user's space allocation per storage tier 560 may also be maintained in a record. Thus, in a tiered storage file system, accounting information may be maintained for each user for each storage tier 560 as well as information for the user aggregated over all storage tiers. This allows an administrator to charge different amounts for the different storage tiers.
  • [0059]
    There are administrative commands to define and report on a charge-back accounting cycle. The reporting cycle is atomically set for all managed objects. A double buffering algorithm and period number are used in each persistent charge-back record to assist in maintaining the current and at least the most recent periods charge-back totals on-the-fly. If the space utilization isn't updated prior to a report being submitted, the correct results will be generated on-demand. Alternatively, a background thread may be used to maintain the charge-back table in real-time.
  • [0060]
    As an example, an administrator might start a new accounting period every day at midnight. The administrator can run the accounting report several times today to get yesterday's accounting information. Tomorrow, the administrator can run the accounting to get today's accounting information.
  • [0061]
    The concept of the minimum and maximum space used by a user may be calculated according to various methods. For example, if the user started the day using 1 GB, then allocated another 1 GB, and then at the end of the day deleted the 1 GB, the minimum would be 1 GB, and the maximum would be 2 GB. Accordingly, the average may be calculated to be 1.5 GB (the average of the minimum and maximum). However, this average is not very meaningful. A more meaningful average may be a weighted average, which takes into account the amount of time that the space was allocated.
  • [0062]
    One example of a weighted average space allocation may be calculated as follows: 12 hours 1 GB + 6 hours 2 GB + 6 hours 1 GB 24 hours = 1.25 GB . Eq . 1
    Equation 1 can be generalized to support an arbitrarynumber of allocations and deallocations in the accounting period. Equation 2 below shows a generalized average calculation, wherein the accounting period corresponds to a history of spacei, i.e., bytes allocated for a duration of timei. Avg = space i time i time i . Eq . 2
  • [0063]
    The problem with Equation 2 is that it requires knowledge of all spacei and timei at the end of the accounting period. However, the average value may be calculated incrementally. Table 1 shows an example of the state of the current accounting buffer 510 shown in FIG. 5 as allocation/deallocation events occur. The current time is relative to the beginning of the accounting period.
    TABLE 1
    Incremental Accounting Updates
    External Events
    Event Current Space Current Accounting Buffer
    Number Time Change Time Space Min Max Avg
    1 0 +1 GB 0 1 GB 1 GB 1 GB 0
    2 12 Hrs +1 GB 12 Hrs 2 GB 1 GB 2 GB   1 GB
    3 18 Hrs −1 GB 18 Hrs 1 GB 1 GB 2 GB 1.33 GB
    4 24 Hrs 0 24 Hrs 1 GB 1 GB 2 GB 1.25 GB

    The table depicts the state of the current accounting buffer after it has been updated as a result of the event. Note that the average lags one event behind; in other words it does not include the 1 GB that was allocated at event number 1 since at the time of the update, that space has been held for no time. The initial value for average at the first event will be the value of space allocation at the beginning of the accounting period (in this example, 0). At the second event the average is updated to include the 1 GB by the formula shown in Equation 3: Avg = Time Avg + ( CurrentTime - Time ) Space CurrentTime . Eq . 3
    In Equation 3, the variables Time, Avg, and Space on the right hand side correspond to the state of the current buffer prior to any updates for this event. Substituting in the values, for the second event results in the calculation: Avg = 0 0 + ( 12 hours - 0 ) 1 GB 12 hours = 1 GB . Eq . 4
    Therefore after the second event, the average is 1 Gb. The value of the average calculated for the third event is calculated incrementally as: Avg = 12 hours 1 GB + ( 18 hours - 12 hours ) 2 GB 18 hours = 1.33 GB . Eq . 5
    Finally at the fourth event, the final average value for the accounting period is computed as: Avg = 18 hours 1.33 GB + ( 24 hours - 18 hours ) 1 GB 24 hours 1.25 GB . Eq . 6
    The fourth event is the first event following the beginning of a new accounting period. At this time, the average will be updated to include the current space used for the remaining time left in the accounting period. This is accomplished by using the end time of the accounting period as the current time when evaluating Equation 3. Once the average has been updated, the contents of the current accounting period buffer 510 as shown in FIG. 5 will be copied to the prior accounting period(s) buffer 542. In addition, after copying the current buffer to the prior accounting period(s) buffer, the current buffer will be initialized by setting the Min, Max, and Avg to Space Allocation 516, and setting the Time 514 to the time that the accounting period began. The fourth event, and its corresponding incremental update to the accounting buffer takes place when either an accounting report is generated or space is allocated or deallocated.
  • [0064]
    FIG. 6 is a flow chart 600 showing a method for calculating a time-weighted average of space allocation according to an embodiment of the present invention. The value representing a maximum space allocation, a minimum space allocation and a time-weighted average for space allocation are maintained 610 The system waits for a change in space allocation 620. If a change is space allocation does not occur 622, the system continues to monitor the space allocation. If the space allocation changes 624, a time-weighted average of space allocation is re-calculated based on the change in space allocation and the length of allocation 630. The system then waits for another change. For example, an event that represents a change may be the end of an accounting period. Thus, at the end of an accounting period, the time-weighted average may be re-calculated as shown in FIG. 6. A determination is made whether the last even was the end of an accounting period 640. If not, the system continues to maintain and update the space allocation data 642. If the last even was the end of an accounting period 644, the process ends. The space allocation data may then be provided to an accounting system.
  • [0065]
    A file system according to an embodiment of the present invention tracks accounting information directly. While some file systems track quota information, such file systems simply allow the current number of bytes used to be determined without being able to providing file system chargeback accounting with real time historical minimum and maximum usage per management object according to an embodiment of the present invention. Further, such file systems must be polled periodically to produce estimates of the minimum, maximum, and average space usage. The benefit of storing and maintaining the accounting information directly within the file system is that the exact minimum, maximum, and average space usage can be determined without the cost of periodically polling the current usage.
  • [0066]
    An accounting period may be defined as including a start date and time and an end date and time, wherein the period is designated as P=<Tstart, Tend>. Over the history of the file system, several accounting periods may occur, denoted as <T0, T1>, <T1, T2>, <T2, T3>, . . . , <TN-1, TN>, and <TN, Tnow>. The end timestamp for one accounting period is the start timestamp for the next accounting period. The last accounting period has an end timestamp of Tnow, which indicates that the accounting period is current. In general, for any completed accounting period Pi, the timestamps are <Ti-1, Ti>. A file system according to an embodiment of the present invention maintains a history of at least two accounting periods, the current period Pnow and at least one previous period. However, a file system according to another embodiment of the present invention may be configured to maintain a longer history, which may include a plurality of previous accounting periods. The history of the accounting periods can be stored persistently using variables that are part of a larger structure that describes the file system.
  • [0067]
    FIG. 7 is a flow chart 700 of the method for managing the content of the buffers relative to changes in accounting periods according to an embodiment of the present invention. In FIG. 7, a determination is made whether the content of the buffer for the previous accounting period is stale 710 (correct?). When updating the accounting information, the start of a new accounting period is noted. The accounting record for each user may have its own local copy of the global index. If the local copy is different than the global, then the accounting record represents a previous accounting period. If the local copy is off by one, the previous buffer is stale 712 and then the minimum, maximum, and average for the new previous buffer is set to the current space usage 720. If the previous buffer is not stale 714, or after the minimum, maximum, and average for the new previous buffer is set to the current space usage, a determination is made whether the current buffer is stale 730. If the local copy of the global index is different from the global index by one or more, then the current buffer is stale 732. The minimum, maximum, and average for the new current buffer is set to the current space usage 740, the average for the new previous buffer is set to include the current space usage for the remainder of the finished accounting period 750. Then, the timestamp is set to end of previous period so that the update time to the start of the accounting period is set 760. If the current buffer is not stale 734, or after the accounting period is set 760, the current buffer is updated 770. Accordingly, when the local index is off by two or more, then both buffers are stale
  • [0068]
    FIG. 8 is a flow chart 800 of the method for providing file system chargeback accounting according to an embodiment of the present invention. Space allocation data for a current accounting period is monitored and maintained in a first storage 810. Space allocation data for a previous accounting period is maintained in a second storage 820 At the end of the current accounting period, a minimum, maximum and average space allocation is calculated for the user during the ended current accounting period 830. The calculated minimum, maximum and average space allocation for the user is moved to the second storage 840. This move may be physical or logical. A new current accounting period is set and space allocation for the new accounting period is monitored and maintained in the first storage 850.
  • [0069]
    FIG. 9 illustrates a system 900 according to an embodiment of the present invention. Embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment containing both hardware and software elements. In a preferred embodiment, the invention is implemented in software, which includes but is not limited to firmware, resident software, microcode, etc. Furthermore, embodiments of the present invention may take the form of a computer program product 990 accessible from a computer-usable or computer-readable medium 968 providing program code for use by or in connection with a computer or any instruction execution system.
  • [0070]
    For the purposes of this description, a computer-usable or computer readable medium 968 can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The medium 968 may be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation medium. Examples of a computer-readable medium include a semiconductor or solid-state memory, magnetic tape, a removable computer diskette, a random access memory (RAM), a read-only memory (ROM), a rigid magnetic disk and an optical disk. Current examples of optical disks include compact disk-read only memory (CD-ROM), compact disk-read/write (CD-R/W) and DVD.
  • [0071]
    A system suitable for storing and/or executing program code will include at least one processor 996 coupled directly or indirectly to memory elements 992 through a system bus 920. The memory elements 992 can include local memory employed during actual execution of the program code, bulk storage, and cache memories which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution.
  • [0072]
    Input/output or I/O devices 930 (including but not limited to keyboards, displays, pointing devices, etc.) can be coupled to the system either directly to the system or through intervening I/O controllers.
  • [0073]
    Network adapters 950 may also be coupled to the system to enable the system to become coupled to other data processing systems 952, remote printers 954 or storage devices 956 through intervening private or public networks 960 (not shown). Modems, cable modem and Ethernet cards are just a few of the currently available types of network adapters.
  • [0074]
    Accordingly, the computer program 990 comprise instructions which, when read and executed by the system 900 of FIG. 9, causes the system 900 to perform the steps necessary to execute the steps or elements of the present invention
  • [0075]
    The foregoing description of the embodiment of the invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. It is intended that the scope of the invention be limited not with this detailed description, but rather by the claims appended hereto.
Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US7243094 *May 30, 2003Jul 10, 2007Softek Storage Solutions CorporationMethod and system for intelligent storage management
US7600007 *May 23, 2000Oct 6, 2009Computer Associates Think, Inc.Method and apparatus for event correlation in service level management (SLM)
US8364564 *Jan 29, 2013International Business Machines CorporationMethod for usage billing in an internet environment
US20030036974 *May 13, 1998Feb 20, 2003Richard AllenApparatus and method for an on demand data delivery system for the preview selection, retrieval and reproduction at a remote location of previously recorded or programmed materials
US20030061362 *Sep 6, 2001Mar 27, 2003Qiu Chaoxin C.Systems and methods for resource management in information storage environments
US20030065617 *Jun 28, 2002Apr 3, 2003Watkins Mark RobertMethod of billing for utilization of a data storage array, and an array controller therefor
US20030110190 *Dec 10, 2001Jun 12, 2003Hitachi, Ltd.Method and system for file space management
US20030115118 *Dec 17, 2001Jun 19, 2003Reinemann Jeffrey K.Resource utilization management
US20040059758 *Sep 20, 2002Mar 25, 2004International Business Machines CorporationMethod and apparatus for optimizing extent size
US20040098383 *May 30, 2003May 20, 2004Nicholas TabellionMethod and system for intelligent storage management
US20040193879 *Aug 8, 2003Sep 30, 2004Hitachi, Ltd.Computer system
US20050115593 *Dec 1, 2003Jun 2, 2005Rudy PublParts washing apparatus
US20050125593 *Jan 5, 2005Jun 9, 2005Yotta Yotta, Inc.Storage virtualization system and methods
US20050144025 *Dec 29, 2003Jun 30, 2005Veritas Operating CorporationUsing technical performance metrics for business and usage analysis and cost allocation
US20060031842 *Sep 8, 2005Feb 9, 2006Steven NeimanSystem and method for dividing computations
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US7778959 *Dec 9, 2005Aug 17, 2010Microsoft CorporationProtecting storages volumes with mock replication
US8006038Oct 2, 2008Aug 23, 2011Microsoft CorporationOptimizing information lifecycle management for fixed storage
US8099484 *Jan 17, 2012Computer Associates Think, Inc.Using mobile agents to perform operations of an information life-cycle management system
US9141288Jul 26, 2012Sep 22, 2015Hewlett-Packard Development Company, L.P.Chargeback based storage recommendations for datacenters
US20070106496 *Nov 9, 2005May 10, 2007Microsoft CorporationAdaptive task framework
US20070136395 *Dec 9, 2005Jun 14, 2007Microsoft CorporationProtecting storage volumes with mock replication
US20070299948 *Jun 27, 2006Dec 27, 2007Pavan Kumar KorleparaUsing Mobile Agents To Perform Operations Of An Information Life-Cycle Management System
US20100088470 *Apr 8, 2010Microsoft CorporationOptimizing information lifecycle management for fixed storage
US20110099095 *Oct 28, 2009Apr 28, 2011Microsoft CorporationProcessing internal use of data-center resources
US20140214755 *Jan 31, 2013Jul 31, 2014Hewlett-Packard Development Company, L.P.Extensible model for it resource chargeback
Classifications
U.S. Classification1/1, 707/E17.01, 707/999.205
International ClassificationG06F17/30
Cooperative ClassificationG06Q40/12
European ClassificationG06Q40/10, G06F17/30F
Legal Events
DateCodeEventDescription
Dec 5, 2005ASAssignment
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HEFFLEY, DANIEL ANSON;KO, WENJENG;SONG, CHENG-CHUNG;REEL/FRAME:017369/0144
Effective date: 20051202
Apr 29, 2010ASAssignment
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION,NEW YO
Free format text: CORRECTIVE ASSIGNMENT TO CORRECT THE TO CORRECT ASSIGNOR NAMES PREVIOUSLY RECORDED ON REEL 017369 FRAME 0144. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT;ASSIGNORS:SEEGER, JAMES JOHN, JR.;YOUNG, JASON CHRISTOPHER;REEL/FRAME:024396/0693
Effective date: 20051202