Management of external memory functioning as virtual cache

Abstract

A method and apparatus for managing the caching of data on an auxiliary memory of a computer. Pages of data may be cached on an auxiliary memory, such as a flash memory, at a virtual level using an identifier that does not involve a physical address of the pages on a memory. Pages may be cached on auxiliary memory that may be removable from the computer, e.g., by unplugging the memory from the computer. Page data may be encrypted and/or compressed on the auxiliary memory. An authentication indicator may be used to verify the accuracy of cached data in the case of an interrupted connection to the auxiliary memory, e.g., as a result of computer power down, hibernation, removal of the memory from the computer, etc.

Description

BACKGROUND

1. Field of Invention

This invention relates to managing the use of auxiliary memory devices as cache memory.

2. Related Art

Improvements to central processing unit (CPU) hardware have, in general, far outpaced advances made in disk drive memory in terms of speed of operation. As a result, a common bottleneck to improved computer performance has been the time needed to retrieve data from disk drive memory for use by the CPU. To help alleviate this problem, computer systems are commonly operated so as to store (or cache) data needed by the CPU in main system memory, e.g., RAM that is typically located on the computer motherboard and has relatively fast read/write times, especially as compared to magnetic disk memory. Thus, needed data can be moved from a disk drive to the main system memory (preferably before the data is needed by the CPU), helping to speed the computer's operation.

SUMMARY OF INVENTION

The inventors have appreciated that the use of auxiliary memory, e.g., memories that may or may not be located on the motherboard, in a computer system as cache memory can improve a computer system's performance, particularly if the auxiliary memory read/write time is faster than the memory(ies) on which most of the system's data is stored (often a disk drive or drives). With the more widespread use of flash memory (e.g., in the form of USB memory sticks or other similar devices), a computer system often has access to memory that operates at a faster rate than a disk drive typically employed to store the bulk of a computer's data. If such auxiliary memory has usable space, the memory can help to speed computer function by caching data from slower memories.

In accordance with an aspect of the invention, auxiliary memories that are removable from the computer system may be effectively used as cache memory. (As used herein, removable means that the memory can be ejected or otherwise disconnected from the computer without causing the computer to crash.) For example, the computer system may use a USB flash memory that a user has plugged into a USB port to cache data for use by the CPU. Management of the cached data may be performed so that even in the event that the auxiliary memory is disconnected or otherwise removed from the system, no loss of data will occur. In one embodiment, only so-called “clean” pages, e.g., blocks of data about 4 kB in size that are not subject to modification, are cached on removable and/or non-removable auxiliary memory. Thus, clean pages may have a backup on another memory, such as an internal disk drive, so that removal of the memory will not cause loss of data. Other pages that may be modified, so-called “dirty” pages, may be cached only on non-removable memory, or may also be stored on removable memory if the system is suitably equipped (e.g., has the ability to ask the user to reconnect a removable memory to the computer if the user wants the system/process to continue running or other ability to recover from the loss of the “dirty” page data).

In another aspect of the invention, the computer system may cache virtual pages on an auxiliary memory or memories. In typical caching techniques, the system caches data at a physical level, i.e., blocks of data identified by physical address in a memory (such as a disk drive) are stored on a second memory (such as a cache memory). The inventors have appreciated that this approach may have disadvantages, including inefficient use of cache memory space. For example, a particular process may only require a relatively small portion of a data block, but the memory manager will typically continue to cache the entire block despite the fact that most of the block will not be used. As another example, if a disk drive is defragmented (which essentially involves the physical rearrangement of data on a disk to free up disk space), a complicated scheme is required to relay to the cache management regarding where data to be cached has been moved on the disk so that appropriate data can be cached.

In one aspect of the invention, data may be cached on a virtual level, e.g., in a way that does not identify a cached page by its physical address on memory device, but rather by a virtual identifier that describes the identity of the cached page at a more abstract level. For example, pages may be virtually identified by a page key that includes a tuple of a process employing the page and a virtual address for the page, or a name or handle of a file that includes the page and a file offset. Other virtual page identifiers may be used as suitable. By identifying a page at a virtual level as opposed to a physical level, more efficient caching may be performed since only those pages expected to be used by the CPU need be cached, potentially avoiding the caching of data that is unlikely to be used. Also, since a process may clearly identify which page(s) are released, e.g., by file or process, the cache management system may rapidly release the corresponding data from the cache.

In another aspect of the invention, data may be cached in one or more auxiliary memories that are removable from the computer system. Such auxiliary memories may include any suitable memory type, such as flash memory, disk drive memory, etc., and may be removable from the computer system by physically disconnecting the memory from the computer system while the computer system continues to otherwise operate normally (e.g., by unplugging a flash memory stick from a computer) and/or by disabling a connection between the computer system and the auxiliary memory (e.g., by disabling a local area network connection via which the computer system communicates with an auxiliary memory). Data may be cached on the auxiliary memory in any suitable way, such as by virtual address as discussed above, and/or by physical address. In accordance with an aspect of the invention, the accuracy of data cached on a removable auxiliary memory may be verified if the auxiliary memory is disconnected from the computer system, and then reconnected.

In one aspect of the invention, a computer system includes a processor adapted to execute instructions to carry out one or more functions, and a main system memory adapted to store a plurality of pages of data that are used by the processor in performing the one or more functions. At least one auxiliary memory may be in communication with the processor and be adapted to store a plurality of pages of data for use by the processor, by the processor directly accessing the auxiliary memory and/or indirectly accessing the auxiliary memory by transferring data to be used by the processor from the auxiliary memory to main system memory. A cache manager (e.g., which may be part of a memory manager in the computer operating system) may control the caching of virtual pages of data for use by the processor on the auxiliary memory and may identify virtual pages using a virtual identifier that is not an address that defines a physical location of the virtual page on a memory. In one embodiment, virtual identifier includes a tuple of file and offset information, or a tuple of process and virtual address information. The auxiliary memory may be removable from the computer, and caching of pages may be managed so that loss of data due to removal of the memory is avoided.

In another aspect of the invention, a method of managing virtual pages cached in auxiliary memory of a computer system includes providing a computer system including a processor adapted to execute instructions to carry out one or more functions, a main system memory adapted to store a plurality of pages of data that are used by the processor in performing the one or more functions, and an auxiliary memory in (direct or indirect) communication with the processor. Virtual pages of data may be cached for use by the processor on the auxiliary memory, and the auxiliary memory may be disconnected from the computer system. In one embodiment, the auxiliary memory may be disconnected from the computer system without loss of the cached data.

In another aspect of the invention, a method of managing pages cached in memory of a computer system includes providing a computer system having a processor adapted to execute instructions to carry out one or more functions, main system memory adapted to store a plurality of pages of data that are used by the processor in performing the one or more functions, and an auxiliary memory in communication with the processor. The auxiliary memory may be adapted to store a plurality of pages of data for use by the processor and be removable from the computer system. Pages of data may be cached for use by the processor on the auxiliary memory, and the auxiliary memory disconnected from the computer system. The auxiliary memory may be reconnected to the computer system, and pages cached on the auxiliary memory may be verified to be consistent with corresponding page data stored in another memory of the computer system.

These and other aspects of the invention will be apparent from the following detailed description and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Aspects of the invention are described with reference to illustrative embodiments and the following drawings in which like numerals reference like elements, and wherein:

FIG. 1 shows a schematic block diagram of a computer arranged in accordance with aspects of the invention; and

FIG. 2 conceptually illustrates an example of managing virtual cache pages in one embodiment.

DETAILED DESCRIPTION

Aspects of the invention are described below with reference to illustrative embodiments. However, it should be appreciated that aspects of the invention are not limited to any of the particular embodiments. For example, embodiments are described below regarding a general purpose programmable computer with one or more auxiliary memories. However, it should be appreciated that aspects of the invention may be employed in any suitable environment, such as where auxiliary memory used as cache memory are linked by a network connection to the computer.

FIG. 1 shows a schematic block diagram of a computer 10 that is arranged in accordance with aspects of the invention. Although in this illustrative embodiment, only selected portions of the computer 10 are identified as being included in the computer 10, this is done for purposes of clarity and not to limit aspects of the invention in any way. For example, the computer 10 may include a display, a keyboard and/or other user input devices (such as a mouse, touch screen or other), communications devices (such as devices suitable for wireless and/or wired communications), as well as any suitable software or other instructions that may be executed by the computer 10 so as perform desired input/output or other functions.

In this illustrative embodiment, the computer 10 includes a central processing unit (CPU or processor) 1 and main system memory 2 that the processor 1 uses for performing its functions. The main system memory 2 may be located on a motherboard with the processor 1 and may have a relatively high read and/or write speed so that data exchanges with the processor 1 can be performed rapidly, e.g., may include RAM that the processor 1 uses to cache data for use in performing various operations. The computer 10 may also include one or more auxiliary memory devices 3, e.g., memories 3 that are not part of the main system memory and may or may not be located on a motherboard with the processor 1. Although the auxiliary memory devices 3 may function as cache memory in accordance with aspects of the invention, these devices 3 are not limited to functioning as a conventional cache memory. For example, the computer 10 may include as auxiliary memory 3 one or more disk drives 3a where a majority of data used by the computer 10 is stored, an optical memory 3b, and a flash memory device 3c. Other auxiliary memory may be provided, such as volatile or non-volatile RAM of any suitable form. Some of the memories 3 may be located within an outer casing of the computer 10, such as the disk drive 3a and the optical memory 3b, whereas others, such as the flash memory 3c may be located outside of the outer casing of the computer 10. Also, some of the memories, such as the disk drive 3a and the optical memory 3b, may not be removable from the computer 10, whereas other of the auxiliary memory, such as the flash memory device 3c, may be removable (e.g., by unplugging the flash memory 3c at a USB port, PCMCIA connection, a memory card reader or other connection to the computer 10). In some embodiments, auxiliary memory 3 may be provided via a network connection to the computer 10 (e.g., by a network server and/or client on the network), and in such cases the auxiliary memory 3 may be removable from the computer 10, i.e., can be physically disconnected from the computer and/or communication with the memory 3 can be disabled.

In accordance with one aspect of the invention, the computer 10 may store cache data on a removable auxiliary memory 3, such as the flash memory 3c. For example, a memory manager 4, which may be part of the operating system for the computer 10, may direct data to be used by the processor 1 that is located on the disk drive 3a to be cached on the flash memory 3c. The faster read and/or write time of the flash memory 3c as compared to the disk drive 3a may help to speed operation of the computer 10 and/or applications executed by the computer 10 (e.g., by reducing operation time perceived by a user), particularly if the main system memory 2 is full or otherwise cannot be used in an efficient way. The memory manager 4 may operate so that pages that are “in use”, i.e., pages that can be modified by an application without operating system intervention, are cached only at the main system memory 2. However, other data may be cached on the auxiliary memory 3. The memory manager 4 may also operate so that only “clean” pages are cached on removable auxiliary memory 3. That is, only pages that are not modified or will not be modified whether by operating system intervention or not are stored on removable auxiliary memory 3. In this way, if the auxiliary memory 3 is removed, the memory manager 4 may resort to the copy of the page on the disk drive 3a or other memory. As for “dirty” pages, i.e., pages that may be or are modified, the memory manager 4 may cache such pages only on non-removable auxiliary memory 3 (or, of course, main system memory 2). In this way, the computer 10 will not lose modified pages due to removal of the auxiliary memory 3.

In accordance with another aspect of the invention, cached pages stored on removable (or non-removable) auxiliary memory 3 may be encrypted. This approach may help to ensure privacy of cached information (such as credit card numbers, bank accounts, passwords, etc.) that is stored on removable auxiliary memory 3, e.g., in case that sensitive information is cached on a removable memory 3 and the memory 3 is removed, taking the sensitive data with it. By encrypting the cached data, it may be difficult or impossible to recover the information from the removable memory 3 without the necessary encryption key or other information needed to recover the encrypted data. One possible way to encrypt cached pages is to generate a random key and encrypt the page(s) using the key and a suitable encryption algorithm. In another example, the key may be salted based on the specific device offset the page is being written to.

In another aspect of the invention, cached pages on auxiliary memory 3 may be compressed. For example, compression ratios of approximately 2:1 or 3:1, depending on the type of data or other factors may be possible, thereby reducing the amount of space needed to cache data. Cached pages that originate from a file, termed “file pages” herein, may experience a compression ratio of 2:1. Cached pages that cannot be shared and that originate from the virtual address space of a process, termed “private pages” herein, may experience a 3:1 compression ratio. As with all other aspects of the invention which may be combined with other aspects in any suitable way, compression of cached pages may be combined with encryption. In fact, in some embodiments, encryption of the cached data may result in compression of the data as well as encryption.

In another aspect of the invention, bytes of data on auxiliary memory 3 may be shared by multiple cached pages. For example, multiple pages containing the same data can be stored in the same location on the auxiliary memory 3 to further optimize space usage.

In another aspect of the invention, a data authenticity indicator may be generated and used to verify that a cached page (or pages) on an auxiliary memory remains unchanged after communication between the computer system and the auxiliary memory is interrupted (e.g., the auxiliary memory has been “removed” from the computer system). Communication with the auxiliary memory 3 may be interrupted in a variety of ways, such as by the computer 10 powering down, crashing, experiencing a power failure, hibernating, etc., a network connection being cut and/or the auxiliary memory 3 being removed from the computer 10. In such cases, the memory manager 4 cannot be assured that the cached data on the auxiliary memory 3 (or data on the source memory from which the cached data was obtained) has not been maliciously changed or otherwise modified without making some kind of verification. In one embodiment, a hash of cached data (or other cryptographic authenticity indicator for the data) on the auxiliary memory 3 may be generated, e.g., after each write of cached data to the auxiliary memory 3, and used in the case of communication interruption to verify that the cached data on the auxiliary memory 3 after reconnection to the computer 10 remains unchanged from before the communication interruption and/or to verify that cached data at the source memory (such as a disk drive) has not changed. The hash may be determined for each page, for each region of the auxiliary memory 3 and/or the entire memory 3. Before entering a sleep state, each memory 3 and/or the memory manager 4 may mark all of the regions of the memory 3 having cached data so that those regions can be hash-verified after the sleep state is terminated. It may be important to ensure that the hash algorithm is cryptographically strong so that one can be reasonably certain that a hash match provides a suitable guarantee that there has been no alteration of data. In another embodiment, the memory manager 4 may maintain a list of changes to data on a source memory so that the memory manager 4 can determine whether there are any differences between the data on the source memory and corresponding cached data on the auxiliary memory 3, and if so, either discard the cached data or update the data so the cached data reflects changes that have occurred. Of course, it will be understood that another technique that may be used to determine differences between cached data on a removed auxiliary memory 3 and data on a source memory is a comparison of the relevant data to determine whether any differences exist. If differences exist, the cached data may be updated using data from the source memory, the source data may be restored (in the case that the cached data on the removed auxiliary memory 3 is determined to be an appropriate data set for use), or the cached data may be discarded.

In another aspect of the invention, data cached on an auxiliary memory may be cached at a virtual level. For example, the memory manager 4 may identify cached pages by a page key that does not include a physical address for the page on a memory and does not include a physical address for a larger block of data on a memory that includes the page. In one embodiment, pages may be identified by a tuple of the page's file identifier and file offset (e.g., for pages originating from a file) or a tuple of a process using the page and the page's virtual address within the process (e.g., for pages originating from a process). The memory manager 4, which may control the caching of data in a main system memory, may include a cache manager 41 that manages the caching of virtual pages on one or more auxiliary memories 3. (Alternately, the cache manager 41 may be a separate component from the memory manager 4 that interacts with the memory manager 4.) In this embodiment, communication between the memory manager 4 and the cache manager 41 uses page keys to identify pages, and pages may be organized in auxiliary memories 3 in respective stores (i.e., an organizational arrangement for pages and information regarding pages). A policy manager (which may or may not be part of the memory manager 4) may make the decisions regarding which pages are placed in stores 31 on auxiliary memories 3 and when. These decisions may be made based on various factors, including store location, store read/write speed, store size, store removability, page identity, page access history, page compressibility, and others. For example, if the memory manager 4 and/or policy manager determines that caching a set of pages on an auxiliary memory 3 having a read/write speed greater than a disk memory would improve performance of the system, the memory manager 4 may direct the pages expected to be used by the processor 1 be copied to the auxiliary memory 3. In one embodiment shown schematically in FIG. 2, when a page is added to a store 31, the memory manager 4 gives the cache manager 41 the page, its key and optionally a set of stores 31 to which the page is to be added. The cache manager 41 may organize stores 31 in an auxiliary memory 3 into regions, which may have a size of about 32 kB to 1 MB when using a 4 kB-8 kB page size. The cache manager 41 (or alternately a store 31) may maintain information regarding the location of each page in each store 31, e.g., expressed as the tuple of a region number and the region offset in the auxiliary memory 3. This information may be stored in a lookup data structure 32, called a PageMap, which may take the form of a B+tree or hash table. The PageMap 32 may include an entry for each page in the store 31, and be searchable by the associated Page key. As can be seen in FIG. 2, the entry for each page may include the Page key tuple (e.g., file/offset or process/virtual address) as well as the location information (e.g., store bitmap) and one or more flags (e.g., priority information, CRC information, etc.).

When retrieving a page from a store 31 (e.g., so that an application can use the page) or removing a page from a store 31, the memory manager 4 provides the page key to the cache manager 41. In response, the cache manager 41 may search the PageMap 32 for the corresponding Page keys, identify the store(s) 31 and location(s) for the page, and provide the page to the memory manager 4. A retrieved page may be placed in main system by the memory manager 4, e.g., so that an application can access the page. If the page is not present, e.g., because an auxiliary memory 3 having the cached page has been removed, the cache manager 41 may indicate a failed operation to the memory manager 4. The memory manager 4 may then resort to other copies of the page. Modified pages or pages that are to be modified generally are removed from stores 31 on auxiliary memory 3 because those copies on the stores 31 will become out of date. The cache manager 41 may remove pages from a store 31 without any input/output being required with the respective auxiliary memory 3, e.g., by updating the PageMap 32. With this arrangement, the memory manager 4 may maintain a relatively small amount of information regarding pages, e.g., the page key and a “present bit” which indicates whether a page is present in a store or not, since the cache manager 41 may maintain sufficient information regarding page location, etc.

Aspects, including embodiments described above, can be implemented in any of numerous ways. For example, the embodiments may be implemented using hardware, software or a combination thereof. When implemented in software, the software code can be executed on any suitable processor or collection of processors, whether provided in a single computer or distributed among multiple computers. It should be appreciated that any component or collection of components that perform the functions described above can be generically considered as one or more controllers that control the above-discussed functions. The one or more controllers can be implemented in numerous ways, such as with dedicated hardware, or with general purpose hardware (e.g., one or more processors) that is programmed using microcode or software to perform the functions recited above.

In this respect, it should be appreciated that one implementation of the embodiments of the present invention comprises at least one computer-readable medium (e.g., a computer memory, a floppy disk, a compact disk, a tape, etc.) encoded with a computer program (i.e., a plurality of instructions), which, when executed on a processor, performs the above-discussed functions of embodiments in accordance with aspects of the present invention. The computer-readable medium can be transportable such that the program stored thereon can be loaded onto any computer environment resource to implement the aspects of the present invention discussed herein. In addition, it should be appreciated that the reference to a computer program which, when executed, performs the above-discussed functions, is not limited to an application program running on a host computer. Rather, the term computer program is used herein in a generic sense to reference any type of computer code (e.g., software or microcode) that can be employed to program a processor to implement the above-discussed aspects of the present invention. It should be appreciated that in accordance with several embodiments of the present invention wherein processes are implemented in a computer readable medium, the computer implemented processes may, during the course of their execution, receive input manually (e.g., from a user).

While aspects of the invention have been described with reference to various illustrative embodiments, the invention is not limited to the embodiments described. Thus, it is evident that many alternatives, modifications, and variations of the embodiments described will be apparent to those skilled in the art. Accordingly, embodiments of the invention as set forth herein are intended to be illustrative, not limiting. Various changes may be made without departing from the invention.

Claims

1. A computing device comprising: a main system memory configured for storing a plurality of pages of data;at least one auxiliary memory configured for storing the plurality of pages of data and for being removable from the computing device; anda memory manager configured for controlling the caching of virtual pages of data on the at least one auxiliary memory, the memory manager further configured for identifying a virtual page of the virtual pages using a page key comprising a tuple, the page key configured for identifying the virtual page as cached in a file based on the tuple comprising the virtual page's file identifier and its file offset information, and further configured for identifying the virtual page as employed by a process based on the tuple comprising the virtual page's process identifier and its virtual address within the process, the memory manager further configured for verifying that virtual pages of data cached on the auxiliary memory are consistent with pages of data stored in the main system memory after the auxiliary memory is removed from the computing device and reconnected to the computing device, where the verifying is based on an analysis of non-time stamp virtual page data on the auxiliary memory.

2. The computing device computer system of claim 1, wherein the tuple of the page key comprises the virtual page's originating file identifier and its originating file offset information corresponding to the virtual page being cached in the file, and wherein the tuple of the page key comprises the virtual page's process identifier and its virtual address within the process corresponding to the virtual page being employed by the process.

3. The computing device of claim 1, wherein the at least one auxiliary memory is removable from the computer system without loss of data corresponding to the virtual pages stored on the at least one auxiliary memory.

4. The computing device of claim 3, wherein the at least one auxiliary memory includes a plurality of memories.

5. The computing device of claim 1, wherein virtual pages stored on the removable at least one auxiliary memory are encrypted.

6. The computing device of claim 1, wherein dirty virtual pages are stored on non-removable auxiliary memories, and clean virtual pages are stored on removable or non-removable auxiliary memories.

7. The computing device of claim 1, wherein the virtual pages are stored in compressed form on the at least one auxiliary memory.

8. The computing device of claim 1, wherein a virtual data authenticity indicator is used by the memory manager to verify that a virtual page on an auxiliary memory remains unchanged after communication between the computer system and the auxiliary memory is interrupted.

9. The computing device of claim 8, wherein the virtual data authenticity indicator includes a hash of the virtual page, and interruption of communication between the computing device and the auxiliary memory is caused by power down of the computing device or removal of the auxiliary memory from the computing device.

10. A method performed on a computing device for managing virtual pages cached in auxiliary memory, the method comprising: providing a main system memory configured for storing a plurality of pages of data, and the auxiliary memory configured for storing the plurality of pages of data and for being removable from the computing device;caching virtual pages of data on the auxiliary memory, a virtual page of the virtual pages being identified by a page key comprising a tuple, the page key configured for identifying the virtual page as cached in a file based on the tuple comprising the virtual page's file identifier and its file offset information, and further configured for identifying the virtual page as employed by a process based on the tuple comprising virtual page's process identifier and its virtual address within the process;disconnecting the auxiliary memory from the computing device;reconnecting the auxiliary memory to the computing device; andverifying that virtual pages cached on the auxiliary memory are consistent with corresponding page data stored in the main system memory, the verifying performed after reconnecting the auxiliary memory to the computing device and based on an analysis of non-time stamp cached virtual page data on the auxiliary memory.

11. The method of claim 10, further comprising caching other virtual pages of data on the auxiliary memory, the virtual pages being identified by a page key that includes a tuple of process and virtual address information.

12. The method of claim 11, further comprising generating a virtual data authenticity indicator used to verify that a virtual page on an auxiliary memory remains unchanged after communication between the computing device and the auxiliary memory is interrupted.

13. The method of claim 10, wherein the disconnecting and the reconnecting are performed without loss of data corresponding to the virtual pages stored on the at least one auxiliary memory.

14. The method of claim 10, wherein virtual pages stored on the auxiliary memory are encrypted.

15. The method of claim 10, wherein the step of caching virtual pages includes: caching dirty virtual pages on non-removable auxiliary memory, andcaching clean virtual pages on removable auxiliary memory.

16. The method of claim 10, wherein the virtual pages are stored in compressed form on the auxiliary memory.

17. The method of claim 10 further comprising: first identifying, in response to the virtual page being cached in the file, the virtual page using the page key based on the tuple of the page key comprising the virtual page's originating file identifier and its originating file offset information; orsecond identifying, in response to the virtual page being employed by the process, the virtual page using the page key based on the tuple of the page key comprising the virtual page's process identifier and its virtual address within the process.

18. A method performed on a computing device for managing pages cached in memory, the method comprising: providing a main system memory configured for storing a plurality of pages of data, and an auxiliary memory configured for storing the plurality of pages of data and for being removable from the computer system;caching virtual pages of data on the auxiliary memory, a virtual page of the virtual pages being identified by a page key comprising a tuple, the page key configured for identifying the virtual page as cached in a file based on the tuple comprising the virtual page's file identifier and file offset information, and further configured for identifying the virtual page as employed by a process based on the tuple comprising the virtual page's process identifier and its virtual address within the process.

19. The method of claim 18 further comprising: disconnecting the auxiliary memory from the computing device;reconnecting the auxiliary memory to computing device to speed computing device operations; andverifying that pages cached on the auxiliary memory are consistent with corresponding page data stored in another memory of the computer system, the verifying performed after reconnecting the auxiliary memory to the computer system and based on an analysis of non-time stamp cached page data on the auxiliary memory, wherein the verifying includes: comparing data cached on the auxiliary memory with the corresponding data stored on the source memory of the computer system from which the data was cached; orassessing whether an authenticity indicator for the data cached on the auxiliary memory indicates a change in the data; ordetermining whether changes to the data cached on the auxiliary memory have occurred based on a list of changes maintained for the auxiliary memory.

20. The method of claim 18 further comprising: first identifying, in response to the virtual page being cached in the file, the virtual page using the page key based on the tuple of the page key comprising the virtual page's originating file identifier and its originating file offset information; andsecond identifying, in response to the virtual page being employed by the process, the virtual page using the page key based on the tuple of the page key comprising the virtual page's process identifier and its virtual address within the process.