INTELLIGENT CACHE MANAGEMENT
An exemplary storage network, storage controller, and methods of operation are disclosed. In one embodiment, a method of managing cache memory in a storage controller comprises receiving, at the storage controller, a cache hint generated by an application executing on a remote processor, wherein the cache hint identifies a memory block managed by the storage controller, and managing a cache memory operation for data associated with the memory block in response to the cache hint received by the storage controller.
The described subject matter relates to electronic computing, and more particularly to systems and methods for intelligent cache management.
BACKGROUNDEffective collection, management, and control of information have become a central component of modern business processes. To this end, many businesses, both large and small, now implement computer-based information management systems.
Data management is an important component of a computer-based information management system. Many users implement storage networks to manage data operations in computer-based information management systems. Storage networks have evolved in computing power and complexity to provide highly reliable, managed storage solutions that may be distributed across a wide geographic area.
In use, various operations are executed against data resident in memory in a storage system. Many storage systems retrieve data that is actively being modified from a permanent storage media and place the data in a cache memory to enhance the speed of executing data operations. Cache memory is a limited resource, and adroit management of cache memory is desirable.
SUMMARYIn one embodiment, a method of managing cache memory in a storage controller comprises receiving, at the storage controller, a cache hint generated by an application executing on a remote processor, wherein the cache hint identifies a memory block managed by the storage controller, and managing a cache memory operation for data associated with the memory block in response to the cache hint received by the storage controller.
Described herein are exemplary storage network architectures and methods for intelligent cache management. The methods described herein may be embodied as logic instructions on a computer-readable medium such as, e.g., firmware executable on a processor. When executed on a processor, the logic instructions cause processor to be programmed as a special-purpose machine that implements the described methods.
Exemplary Network ArchitectureA plurality of logical disks (also called logical units or LUs) 112a, 112b may be allocated within storage pool 110. Each LU 112a, 112b comprises a contiguous range of logical addresses that can be addressed by host devices 120, 122, 124 and 128 by mapping requests from the connection protocol used by the host device to the uniquely identified LU 112. As used herein, the term “host” comprises a computing system(s) that utilize storage on its own behalf, or on behalf of systems coupled to the host. For example, a host may be a supercomputer processing large databases or a transaction processing server maintaining transaction records. Alternatively, a host may be a file server on a local area network (LAN) or wide area network (WAN) that provides storage services for an enterprise. A file server may comprise one or more disk controllers and/or RAID controllers configured to manage multiple disk drives. A host connects to a storage network via a communication connection such as, e.g., a Fibre Channel (FC) connection.
A host such as server 128 may provide services to other computing or data processing systems or devices. For example, client computer 126 may access storage pool 110 via a host such as server 128. Server 128 may provide file services to client 126, and may provide other services such as transaction processing services, email services, etc. Hence, client device 126 may or may not directly use the storage consumed by host 128.
Devices such as wireless device 120, and computers 122, 124, which are also hosts, may logically couple directly to LUs 112a, 112b. Hosts 120-128 may couple to multiple LUs 112a, 112b, and LUs 112a, 112b may be shared among multiple hosts. Each of the devices shown in
Client computers 214a, 214b, 214c may access storage cells 210a, 210b, 210c through a host, such as servers 216, 220, 230. Clients 214a, 214b, 214c may be connected to file server 216 directly, or via a network 218 such as a Local Area Network (LAN) or a Wide Area Network (WAN). The number of storage cells 210a, 210b, 210c that can be included in any storage network is limited primarily by the connectivity implemented in the communication network 212. A switching fabric comprising a single FC switch can interconnect 256 or more ports, providing a possibility of hundreds of storage cells 210a, 210b, 210c in a single storage network.
Hundreds or even thousands of host computers 216, 220 may connect to storage network 200 to access data stored in storage cells 210a, 210b, 210c. Hosts 216, 220 may be embodied as server computers.
Computing device 330 further includes a hard disk drive 344 for reading from and writing to a hard disk (not shown), and may include a magnetic disk drive 346 for reading from and writing to a removable magnetic disk 348, and an optical disk drive 350 for reading from or writing to a removable optical disk 352 such as a CD ROM or other optical media. The hard disk drive 344, magnetic disk drive 346, and optical disk drive 350 are connected to the bus 336 by a SCSI interface 354 or some other appropriate interface. The drives and their associated computer-readable media provide nonvolatile storage of computer-readable instructions, data structures, program modules and other data for computing device 330. Although the exemplary environment described herein employs a hard disk, a removable magnetic disk 348 and a removable optical disk 352, other types of computer-readable media such as magnetic cassettes, flash memory cards, digital video disks, random access memories (RAMs), read only memories (ROMs), and the like, may also be used in the exemplary operating environment.
A number of program modules may be stored on the hard disk 344, magnetic disk 348, optical disk 352, ROM 338, or RAM 340, including an operating system 358, one or more application programs 360, other program modules 362, and program data 364. A user may enter commands and information into computing device 330 through input devices such as a keyboard 366 and a pointing device 368. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner, or the like. These and other input devices are connected to the processing unit 332 through an interface 370 that is coupled to the bus 336. A monitor 372 or other type of display device is also connected to the bus 336 via an interface, such as a video adapter 374.
Computing device 330 may operate in a networked environment using logical connections to one or more remote computers, such as a remote computer 376. The remote computer 376 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to computing device 330, although only a memory storage device 378 has been illustrated in
When used in a LAN networking environment, computing device 330 is connected to the local network 380 through a network interface or adapter 384. When used in a WAN networking environment, computing device 330 typically includes a modem 386 or other means for establishing communications over the wide area network 382, such as the Internet. The modem 386, which may be internal or external, is connected to the bus 336 via a serial port interface 356. In a networked environment, program modules depicted relative to the computing device 330, or portions thereof, may be stored in the remote memory storage device. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.
Hosts 216, 220 may include host adapter hardware and software to enable a connection to communication network 212. The connection to communication network 212 may be through an optical coupling or more conventional conductive cabling depending on the bandwidth requirements. A host adapter may be implemented as a plug-in card on computing device 330. Hosts 216, 220 may implement any number of host adapters to provide as many connections to communication network 212 as the hardware and software support.
Generally, the data processors of computing device 330 are programmed by means of instructions stored at different times in the various computer-readable storage media of the computer. Programs and operating systems distributed, for example, on floppy disks, CD-ROMs, or electronically, and are installed or loaded into the secondary memory of a computer. At execution, the programs are loaded at least partially into the computer's primary electronic memory.
Each NSC 410a, 410b further includes a communication port 428a, 428b that enables a communication connection 438 between the NSCs 410a, 410b. The communication connection 438 may be implemented as a FC point-to-point connection, or pursuant to any other suitable communication protocol.
In an exemplary implementation, NSCs 410a, 410b further include a plurality of Fiber Channel Arbitrated Loop (FCAL) ports 420a-426a, 420b-426b that implement an FCAL communication connection with a plurality of storage devices, e.g., arrays of disk drives 440, 442. While the illustrated embodiment implement FCAL connections with the arrays of disk drives 440, 442, it will be understood that the communication connection with arrays of disk drives 440, 442 may be implemented using other communication protocols. For example, rather than an FCAL configuration, a FC switching fabric may be used.
Exemplary OperationsHaving described various components of an exemplary storage network, attention is now directed to operations of the storage network 200 and components thereof.
In operation, application software executing on one or more client computing devices such as computing device 126 perform functions that generate requests which are directed to a host computer such as host computer 128. In response to the request(s), the host computer 128 transmits a data request to a component of the storage network 200. Typically, the request is transmitted to one or more NSC such as NSC 410a, 410b, which executes the data operation against data on a disk. The NSC may retrieve data from storage on permanent media such as one or more of the disks 440, 442 into cache memory such as memory 418a, 418b.
To facilitate efficient management of cache memory resources, a controller may be configured to receive cache management hints generated by higher-level software applications and to manage the cache resource in response to the received cache management hints. In one implementation, a cache management hint may be embodied as a message that provides a suggestion to the NSC such as NSC 410a, 410b regarding how the data passed by the application should be managed in cache memory. The NSC may be configured to receive and to respond to cache management hints. In alternate implementations, cache management instructions may be received and processed by a different processor communicatively connected to an NSC, and appropriate cache management instructions may be transmitted from the processor to the NSC.
Each module 510-540 may generate one or more cache hints, which may be transmitted directly to the storage controller 550. For example, application software 510 may generate one or more cache hints which may be transmitted directly to the storage controller 550. Similarly, one or more of the database management software 520, file management system 530 and/or operating system 550 may generate one or more cache hints which may be transmitted directly to the storage controller 550.
Alternatively, each module in the stack 500 may transmit one or more cache hints to the next module in the stack. In one embodiment the respective modules in the stack 500 pass cache hints to the next module in the stack without performing any analysis on the received cache hints. In alternate embodiments one or more of the modules 510-540 may be configured to analyze cache hints received from high-level modules in the stack and to generate a response in the event that the analysis indicates that the cache hint may cause an error.
By way of example, database management software module 520 may analyze cache hints received from one or more application modules 510 to determine whether the cache hints may invoke operations that conflict with database management software 520. Similarly, file management system module 530 may analyze cache hints received from one or more database management software modules 520 to determine whether cache hints may invoke operations that interfere with file management system module 530. Operating system 540 may analyze cache hints received from one or more file management system modules 530 to determine whether cache hints may invoke operations that interfere with operating system 540.
When one or more of the software modules 510-540 detects a conflict the software module may perform one or more remedial operations. By way of example the module may generate an error warning that may be passed back up through the stack 500 and presented to a user of the system, or to an administrator. The error warning may also be stored in a log for subsequent analysis. The module may also be configured to modify the cache hint in a fashion that reconciles a potential conflict indicated by the analysis, or to cancel the cache hint entirely.
In one embodiment storage controller 550 is configured to receive cache hints and to manage cache operations in response to the cache hints.
Referring to
In one embodiment a Write Through Cache hint may identify data associated with the operation by logical unit number (LUN). In alternate embodiments data may be identified by a logical or physical memory address associated with the data. The particular mechanism by which the data is identified is not important.
It is not necessary to enforce strict symmetry between the data associated with Pin Cache and Unpin Cache operations. A data block may be “pinned” to cache using a Pin Cache hint, and may be “unpinned” in a series of sub-blocks at different points in time. The storage controller may implement additional routines associated with Pin Cache hints to manage cache resources effectively. For example, a Pin Cache hint may be subject to a time threshold governed by the storage controller, so that data “pinned” to cache is automatically “unpinned” when the time threshold expires. Alternatively, the storage controller may impose a storage capacity threshold associated with “pinned” data, so that when the amount of “pinned” data exceeds a storage space threshold some or all of the pinned data is unpinned from cache.
In response to the request, at operation 920, the storage controller pre-fetches the identified blocks of data from the permanent storage media and stores the data in cache memory, and returns an acknowledgment to the host computer at operation 930.
In one embodiment data blocks may be assigned numeric priority values that fall within a range of values. The storage controller may maintain a queue of data blocks that are to be de-staged to permanent storage media, and data blocks may be positioned in the queue in accordance with their respective priority values assigned. Thus, high-priority data blocks may be placed near the front of the queue, while low priority blocks may be placed near the end of the queue.
The controller can manage multiple working sets per group, and multiple groups may access a working set.
Referring to
The operations set forth in
Claims
1-28. (canceled)
29. A storage device, comprising:
- at least one disk drive;
- a controller to manage data input/output operations between the disk drive and a remote processor; and
- a cache memory module communicatively connected to the controller,
- wherein the controller is configured to receive a Pin cache hint generated by an application executing on a remote computing device and the controller holds one or more data blocks identified in the Pin cache hint based on the cache hint received by the controller.
30. The storage device of claim 29, wherein the controller holds one or more data blocks identified in the Pin cache hint until an Unpin cache hint identifying one or more data blocks identified in the Pin cache hint is received.
31. The storage device of claim 29, wherein the controller holds the one or more data blocks identified in the Pin cache hint until a time threshold expires.
32. The storage device of claim 29, wherein the controller holds the one or more data blocks identified in the Pin cache hint until a capacity threshold is reached.
33. The storage device of claim 29, wherein the controller receives the Pin cache hint in a header of a data operation request.
34. The storage device of claim 29, wherein the remote computing device comprises a plurality of modules in a software stack executing on the remote computing device wherein the Pin cache hint is generated by a first module in the software stack and analyzed by a second module in the program stack to determine whether the Pin cache hint conflicts with the second module.
35. The storage device of claim 34, wherein the plurality of modules of the program stack comprise at least two of an application software, a database management software, a file management system, and an operating system.
36. The storage device of claim 29, wherein the controller is configured to acknowledge completion of a cache operation to the remote computing device.
37. A storage device, comprising:
- at least one disk drive;
- a controller to manage data input/output operations between the disk drive and a remote processor; and
- a cache memory module communicatively connected to the controller,
- wherein the controller is configured to receive a Working Set cache hint generated by an application executing on a remote computing device, the Working Set cache hint identifying one or more data blocks; and
- the controller caches the one or more data blocks identified in the Working Set cache hint together as a group.
38. The storage device of claim 37, wherein the controller receives the Working Set cache hint in a header of a data operation request.
39. The storage device of claim 37, wherein the remote computing device comprises a plurality of modules in a software stack executing on the remote computing device wherein the Working Set cache hint is generated by a first module in the software stack and analyzed by a second module in the program stack to determine whether the Working Set cache hint conflicts with the second module.
40. The storage device of claim 39, wherein the plurality of modules of the program stack comprise at least two of an application software, a database management software, a file management system, and an operating system.
41. The storage device of claim 37, wherein the controller is configured to acknowledge completion of a cache operation to the remote computing device.
42. A storage device, comprising:
- at least one disk drive;
- a controller to manage data input/output operations between the disk drive and a remote processor, wherein the controller maintains a queue of data blocks for executing cache operations; and
- a cache memory module communicatively connected to the controller;
- wherein the controller is configured to receive a Priority cache hint that associates a priority level for cache operations associated with one or more data blocks identified in the cache hint, and the storage controller organizes data blocks in the queue in accordance with the priority level.
43. The storage device of claim 42, wherein the Priority cache hint includes a priority parameter associated with a data block identified in the cache hint, and the data block is de-staged to a permanent storage media of the storage device in accordance with the priority parameter associated with the data block.
44. The storage device of claim 43, wherein the controller maintains a de-staging queue comprising the data block, and the data block is positioned in the de-staging queue in accordance with the priority parameter.
45. The storage device of claim 42, wherein the controller receives the Priority cache hint in a header of a data operation request.
46. The storage device of claim 42, wherein the remote computing device comprises a plurality of modules in a software stack executing on the remote computing device wherein the Priority cache hint is generated by a first module in the software stack and analyzed by a second module in the program stack to determine whether the Priority cache hint conflicts with the second module.
47. The storage device of claim 46, wherein the plurality of modules of the program stack comprise at least two of an application software, a database management software, a file management system, and an operating system.
48. The storage device of claim 42, wherein the controller is configured to acknowledge completion of a cache operation to the remote computing device.
Type: Application
Filed: Oct 21, 2011
Publication Date: Feb 16, 2012
Inventor: Curt Kolovson (Redwood City, CA)
Application Number: 13/278,541
International Classification: G06F 12/08 (20060101);