Abstract: Aspects of the present disclosure relate to systems and methods for improving performance of a partial cache collapse by a processing device. Certain embodiments provide a method for performing a partial cache collapse procedure, the method including: counting a number of cache lines that satisfy an eviction criteria based on a deterministic cache eviction policy in each cache way of a group of cache ways; selecting at least one cache way from the group for collapse, based on its corresponding number of cache lines that satisfy the eviction criteria; and performing the partial cache collapse procedure based on the at least one cache way selected from the group for collapse.

Abstract: A method includes storing a function representing a set of data elements stored in a backing memory and, in response to a first memory read request for a first data element of the set of data elements, calculating a function result representing the first data element based on the function.

Abstract: Processing a read request to read metadata from an entry of a metadata page may include: determining whether the metadata page is cached; responsive to determining the metadata page is cached, obtaining the first metadata from the cached metadata page; responsive to determining the metadata page is not cached, determining whether the requested metadata is in a metadata log of metadata changes stored in a volatile memory; and responsive to determining the metadata is the metadata log of metadata changes stored in the volatile memory, obtaining the requested metadata from the metadata log. Processing a write request that overwrites an existing value of a metadata page with an updated value may include: recording a metadata change in the metadata log that indicates to update the metadata page with the updated value; and performing additional processing during destaging that uses the existing value prior to overwriting it with the updated value.

Abstract: Systems and techniques for dynamic selection of policy that determines whether copies of shared cache lines in a processor core complex are to be stored and maintained in a level 3 (L3) cache of the processor core complex are based on one or more cache line sharing parameters or based on a counter that tracks L3 cache misses and cache-to-cache (C2C) transfers in the processor core complex, according to various embodiments. Shared cache lines are shared between processor cores or between threads. By comparing either the cache line sharing parameters or the counter to corresponding thresholds, a policy is set which defines whether copies of shared cache lines at such indices are to be retained in the L3 cache.

Abstract: A query for opaque objects stored within a data store and not currently cached within a cache is received from an application. The query to the data store is passed to the data store, and a handle to memory location is received from the data store at which the data store has temporarily stored a message including the opaque objects. The opaque objects are added to the cache by treating the memory location as a cache entry for the opaque objects. Cache metadata for the cache entry is generated, and along with the handle is stored within a metadata cache entry of a metadata cache structure separate from the message. The handle and the cache metadata can be returned to the application, where the cache metadata can be returned as an opaque context of the cache entry.

Abstract: A method, device, and system for controlling a data read/write command in an NVMe over fabric architecture. In the method provided in the embodiments of the present disclosure, a data processing unit receives a control command sent by a control device, the data processing unit divides a storage space of a buffer unit into at least two storage spaces according to the control command sent by the control device, and establishes a correspondence between the at least two storage spaces and command queues, and after receiving a first data read/write command that is in a first command queue and that is sent by the control device, the data processing unit buffers, in a storage space that is of the buffer unit and that is corresponding to the first command queue, data to be transmitted according to the first data read/write command.

Abstract: Embodiments of the present application provide a method and apparatus for updating search cache, which relate to the technical field of multimedia.

Abstract: Techniques described herein relate to systems and methods of data storage, and more particularly to providing layering of file system functionality on an object interface. In certain embodiments, file system functionality may be layered on cloud object interfaces to provide cloud-based storage while allowing for functionality expected from a legacy applications. For instance, POSIX interfaces and semantics may be layered on cloud-based storage, while providing access to data in a manner consistent with file-based access with data organization in name hierarchies. Various embodiments also may provide for memory mapping of data so that memory map changes are reflected in persistent storage while ensuring consistency between memory map changes and writes. For example, by transforming a ZFS file system disk-based storage into ZFS cloud-based storage, the ZFS file system gains the elastic nature of cloud storage.

Abstract: A method includes monitoring one or more metrics for each of a plurality of cache users sharing a cache, and assigning each of the plurality of cache users to one of a plurality of groups based on the monitored one or more metrics.

Abstract: Database environments may choose to schedule complex analytics processing to be performed by specialized processing environments by caching source datasets or other data needed for the analytics and then outputting results back to customer datasets. It is complex to schedule user database operations, such as running dataflows, recipes, scripts, rules, or the like that may rely on output from the analytics, if the user database operations are on one schedule, while the analytics is on another schedule. User/source datasets may become out of sync and one or both environments may operate on stale data. One way to resolve this problem is to define triggers that, for example, monitor for changes to datasets (or other items of interest) by analytics or other activity and automatically run dataflows, recipes, or the like that are related to the changed datasets (or other items of interest).

Abstract: Embodiments operate a multi-tenant cloud system. At a first data center, embodiments authenticate a first client corresponding to a first tenant ID and store resources that correspond to the first client, the first data center in communication with a second data center that is configured to authenticate the first client and replicate the resources. The first data center receives an Application Programming Interface (“API”) request for the first client corresponding to a change to the resources, and generates a change log and corresponding change event message in response to the API request. Embodiments compute a first hash corresponding to the first tenant ID of the change log to determine a first partition of a first queue at the first data center. The first data center pushes the change event message to the second data center via an API call.

Abstract: A method of managing a storage system comprises detecting a reference to a first page in the storage system. The method also comprises creating a first candidate block for the first page based on the detecting. The first candidate block may comprise a continuous series of pages that begins with the first page. The method also comprises monitoring subsequent references to pages within the first candidate block. The method also comprises determining that the first candidate block meets a first set of hot-block requirements. The method also comprises relocating the first candidate block to a hot-block space in a buffer pool based on the determining, resulting in a first hot block.

Abstract: Embodiments of the present disclosure relate to cache memory management. Based on anticipated input/output (I/O) workloads, at least one or more of: sizes of one or more mirrored and un-mirrored caches of global memory and their respective cache slot pools are dynamically balanced. Each of the mirrored/unmirrored caches can be segmented into one or more cache pools, each having slots of a distinct size. Cache pool can be assigned an amount of the one or more cache slots of the distinct size based on the anticipated I/O workloads. Cache pools can be further assigned the amount of distinctly sized cache slots based on expected service levels (SLs) of a customer. Cache pools can also be assigned the amount of the distinctly sized cache slots based on one or more of predicted I/O request sizes and predicted frequencies of different I/O request sizes of the anticipated I/O workloads.

Abstract: One example method includes determining a fall through threshold value for a cache, computing a length ‘s’ of a sequence that is close to LRU eviction, and the length ‘s’ is computed when a current fall through metric value is greater than the fall through threshold value, when the sequence length ‘s’ is greater than a predetermined threshold length ‘k,’ performing a first shift of an LRU position to define a protected queue of the cache, initializing a counter with a value of ‘r’, decrementing the counter each time a requested page is determined to be included in the protected queue, until ‘r’=0, and performing a second shift of the LRU position.

Abstract: Systems and methods are disclosed including a first memory device, a second memory device coupled to the first memory device, where the second memory device has a lower access latency than the first memory device and acts as a cache for the first memory device. A processing device operatively coupled to the first and second memory devices can track access statistics of segments of data stored at the second memory device, the segments having a first granularity, and determine to update, based on the access statistics, a segment of data stored at the second memory device from the first granularity to a second granularity. The processing device can further retrieve additional data associated with the segment of data from the first memory device and store the additional data at the second memory device to form a new segment having the second granularity.

Abstract: A mechanism is described for facilitating dynamic cache allocation in computing devices in computing devices. A method of embodiments, as described herein, includes facilitating monitoring one or more bandwidth consumptions of one or more clients accessing a cache associated with a processor; computing one or more bandwidth requirements of the one or more clients based on the one or more bandwidth consumptions; and allocating one or more portions of the cache to the one or more clients in accordance with the one or more bandwidth requirements.

Abstract: A computer system is provided. The computer system can include a memory, a network interface, and at least one processor coupled to the memory and the network interface. The at least one processor can be configured to identify a file to provide to a computing device; predict a geolocation at which the computing device is to request access to the file; predict a network bandwidth to be available to the computing device at the geolocation; determine, based on the file and the network bandwidth, a first portion of the file to store in a cache of the computing device; and download, via the network interface, the first portion of the file to the cache.

Abstract: An appliance management system manages information about appliances. The system includes an information collection system and an information transmission apparatus. The information collection system includes a database storing the information about the appliances, and first and second reception processing units that write information about the appliances that has been received to the database. The information transmission apparatus includes a transmission unit that transmits the information about the appliances to the first or second reception processing unit via a communication line, and a transmission information creation unit that creates transmission information as the information about the appliances. The transmission information creation unit creates as the transmission information to be sent to the first and second reception processing units, first and second transmission information from first and second type information about the appliances, respectively.

Abstract: A method for dynamically adjusting cache memory partition sizes within a storage system includes computing a read hit ratio for data accessed in each cache partition and an average read hit ratio for all the cache partitions over a time interval. The cache memory includes a higher performance portion (DRAM) and lower performance portion (SCM). The method increases or decreases the partition size for each cache partition by comparing the read hit ratio for the partition to the average read hit ratio for all the partitions. Each cache partition includes maximum and minimum partition sizes, and read hit and read access counters. The SCM portion of the cache memory includes cache partitions reserved for storing data of a specific type, or data used for a specific purpose or with a specific software application. A corresponding storage controller and computer program product are also disclosed.

Abstract: A data storage array is configured for m-way resiliency across a first plurality of storage nodes. The m-way resiliency causes the data storage array to direct each top-level write to at least m storage nodes within the first plurality, for committing data to a corresponding capacity region allocated on each storage node to which each write operation is directed. Based on the data storage array being configured for m-way resiliency, an extra-resilient cache is allocated across a second plurality of storage nodes comprising at least s storage nodes (where s>m), including allocating a corresponding cache region on each of the second plurality for use by the extra-resilient cache. Based on determining that a particular top-level write has not been acknowledged by at least n of the first plurality of storage nodes (where n m), the particular top-level write is redirected to the extra-resilient cache.

Abstract: The invention relates to a computer-implemented method, a corresponding a computer program product and a corresponding apparatus for distributing cached content in a network, the computer-implemented method comprising: collecting statistics regarding requests made and paths taken by the requests from source nodes to server nodes via intermediate nodes, the source nodes, intermediate nodes, and server nodes interconnected by edges having queues with respective queue sizes associated therewith, the requests including indications of content items to be retrieved; storing the content items at the server nodes; caching, by the intermediate nodes, the content items up to a caching capacity; and performing caching decisions that determine which of the content items are to be cached at which of the intermediate nodes, based upon costs that are monotonic, non-decreasing functions of the sizes of the queues.

Abstract: Systems, methods, apparatuses, and computer readable media may be configured for improved predictive content caching. A system may determine a value that is a function of one or more rates at which a portion of a content item is being consumed and based on this value, may also calculate a projected position after a predetermined time period. By comparing the projected position to a dynamically adjustable threshold position for requesting a new portion of the content item, a determination may be made as to when to retrieve and/or cache a new portion of the content item.

Abstract: Methods and systems for implementing independent time in a hosted operating environment are disclosed. The hosted, or guest, operating environment, can be seeded with a guest time value by a guest operating environment manager that maintains a time delta between a host clock time and an enterprise time. The guest operating environment can subsequently manage its guest clock from the guest time value. If the guest operating environment is halted, the guest operating environment manager can manage correspondence between the host clock time and the enterprise time by periodically assessing divergence between actual and expected values of the host clock time.

Abstract: Memory transactions can be tagged with a partition identifier selected depending on which software execution environment caused the memory transaction to be issued. A memory system component can control allocation of resources for handling the memory transaction or manage contention for said resources depending on a selected set of memory system component parameters selected depending on the partition identifier specified by the memory transaction, or can control, depending on the partition identifier specified by the memory transaction, whether performance monitoring data is updated in response to the memory transaction. Page table walk memory transactions may be assigned a different partition identifier to the partition identifier assigned to the corresponding data/instruction access memory transaction.

Abstract: System and methods are disclosed include a memory device and a processing device coupled to the memory device. The processing device can determine an amount of valid blocks in a memory device of a memory sub-system. The processing device can then determine a surplus amount of valid blocks on the memory device based on the amount of valid blocks. The processing device can then configure a size of a cache of the memory device based on the surplus amount of valid blocks.

Abstract: There is provided a method in a first device of a cellular communication system, the method comprising: acquiring a first value of a performance indicator; causing a transmission of management plane performance data to a second device of the cellular communication system, said performance data comprising said first value; acquiring a second value of the performance indicator; and preventing a transmission of the second value if the second value is substantially equal to the first value.

Abstract: Disclosed are embodiments for migrating a virtual machine (VM) from a source host to a destination host while the virtual machine is running on the destination host. The system includes an RDMA facility connected between the source and destination hosts and a device coupled to a local memory, the local memory being responsible for memory pages of the VM instead of the source host. The device is configured to copy pages of the VM to the destination host and to maintain correct operation of the VM by monitoring coherence events, such as a cache miss, caused by the virtual machine running on the destination host. The device services these cache misses using the RDMA facility and copies the cache line satisfying the cache miss to the CPU running the VM. The device also tracks the cache misses to create an access pattern that it uses to predict future cache misses.

Abstract: Techniques for performing cache management includes partitioning entries of a hash table into buckets, wherein each of the buckets includes a portion of the entries of the hash table, configuring a cache, wherein the configuring includes allocating a section of the cache for exclusive use by each bucket, and performing first processing that stores a data block in the cache. The first processing includes determining a hash value for a data block, selecting, in accordance with the hash value, a first bucket of the plurality of buckets, wherein a first section of the cache is used exclusively for storing cached data blocks of the first bucket, storing metadata used in connection with caching the data block in a first entry of the first bucket, and storing the data block in a first cache location of the first section of the cache.

Abstract: A method for cache coherency in a reconfigurable cache architecture is provided. The method includes receiving a memory access command, wherein the memory access command includes at least an address of a memory to access; determining at least one access parameter based on the memory access command; and determining a target cache bin for serving the memory access command based in part on the at least one access parameter and the address.

Abstract: A technique is introduced for intercepting memory calls from a user-space application and applying an allocation policy to determine whether such calls are handled using volatile memory such as dynamic random-access memory (DRAM) or persistent memory (PMEM). In an example embodiment, memory calls from an application are intercepted by a memory allocation capture library. Such calls may be to a memory function such as malloc( ) and may be configured to cause a portion of DRAM to be allocated to the application to process a task. The memory allocation capture library then determines whether the intercepted call satisfies capture criteria associated with an allocation policy. If the intercepted call does satisfy the capture criteria, the call is processed to cause a portion of PMEM to be allocated to the application instead of DRAM.

Abstract: A method, computer program product, and computing system for writing, from a first node to a second node, a first portion of data from a memory pool in the first node defined by, at least in part, a first pointer. One or more input/output (IO) operations may be received while writing the first portion of data to the second node. Data from the one or more IO operations may be stored within the memory pool after the first pointer.

Abstract: Embodiments of the present disclosure relate to an apparatus comprising a memory and at least one processor. The at least one processor is configured to: analyze input/output (I/O) operations received by a storage system; dynamically predict anticipated I/O operations of the storage system based on the analysis; and dynamically control a size of a local cache of the storage system based on the anticipated I/O operations.

Abstract: In one example, a method may include receiving a write operation corresponding to a portion of a data chunk stored at a first storage location in a write-in-place file system. The write-in-place file system may include encoded data chunks and unencoded data chunks. The method may include determining whether the data chunk is an encoded data chunk based on metadata associated with the data chunk, modifying the data chunk based on the write operation, and selectively performing a redirect-on-write operation on the modified data chunk based on the determination.

Abstract: Database environments may choose to schedule complex analytics processing to be performed by specialized processing environments by caching source datasets or other data needed for the analytics and then outputting results back to customer datasets. It is complex to schedule user database operations, such as running dataflows, recipes, scripts, rules, or the like that may rely on output from the analytics, if the user database operations are on one schedule, while the analytics is on another schedule. User/source datasets may become out of sync and one or both environments may operate on stale data. One way to resolve this problem is to define triggers that, for example, monitor for changes to datasets (or other items of interest) by analytics or other activity and automatically run dataflows, recipes, or the like that are related to the changed datasets (or other items of interest).

Abstract: Techniques for cache management may include: partitioning a cache into buckets of cache pages, wherein each bucket has an associated cache page size and each bucket includes cache pages of the associated cache page size for that bucket, wherein the cache includes compressed pages of data and uncompressed pages of data; and performing processing that stores a first page of data in the cache. The processing may include storing the first page of data in a first cache page of a selected bucket having a first associated cache page size determined in accordance with a first compressed size of the first page of data. The cache may be repartitioned among the buckets based on associated access frequencies of the buckets of cache pages.

Abstract: A computational device receives indications of a minimum retention time and a maximum retention time in cache for a first plurality of tracks, wherein no indications of a minimum retention time or a maximum retention time in the cache are received for a second plurality of tracks. A cache management application demotes a track of the first plurality of tracks from the cache, in response to determining that the track is a least recently used (LRU) track in a LRU list of tracks in the cache and the track has been in the cache for a time that exceeds the minimum retention time. The cache management application demotes the track of the first plurality of tracks, in response to determining that the track has been in the cache for a time that exceeds the maximum retention time.

Abstract: Techniques are provided for computer memory mapping and allocation. In an example, a virtual memory address space is divided into an active half and a passive half. Processors make memory allocations to their respective portions of the active half until one processor has made a determined number of allocations. When that occurs, and when all memory in the passive half that has been allocated has been returned, then the active and passive halves are switched, and all processors are switched to making allocations in the newly-active half.

Abstract: A computer system includes a cache and processor. The cache includes a plurality of data compartments configured to store data. The data compartments are arranged as a plurality of data rows and a plurality of data columns. Each data row is defined by an addressable index. The processor is in signal communication with the cache, and is configured to operate in a full cache purge mode and a selective cache purge mode. In response to invoking one or both of the full cache purge mode and the selective cache purge mode, the processor performs a pipe pass on a selected addressable index to determine a number of valid compartments and a number of invalid compartments, and performs an eviction operation on the valid compartments while skipping the eviction operation on the invalid compartments.

Abstract: A list of a first type of tracks in a cache is generated. A list of a second type of tracks in the cache is generated, wherein I/O operations are completed relatively faster to the first type of tracks than to the second type of tracks. A determination is made as to whether to demote a track from the list of the first type of tracks or from the list of the second type of tracks.

Abstract: In an example, an apparatus comprises a plurality of compute engines; and logic, at least partially including hardware logic, to detect a cache line conflict in a last-level cache (LLC) communicatively coupled to the plurality of compute engines; and implement context-based eviction policy to determine a cache way in the cache to evict in order to resolve the cache line conflict. Other embodiments are also disclosed and claimed.

Abstract: A method, computer program product, and computer system for executing an automatic recovery of log metadata. A secondary storage processor may request one or more log metadata buffer values from a first buffer used by a primary storage processor. The secondary storage processor may update one or more log metadata buffer values from a second buffer used by the secondary storage processor.

Abstract: Techniques described herein generally include methods and systems related to cache partitioning in a chip multiprocessor. Cache-partitioning for a single thread or application between multiple data sources improves energy or latency efficiency of a chip multiprocessor by exploiting variations in energy cost and latency cost of the multiple data sources. Partition sizes for each data source may be selected using an optimization algorithm that minimizes or otherwise reduces latencies or energy consumption associated with cache misses.

Abstract: Systems and methods provide for optimizing utilization of an Address Translation Cache (ATC). A network interface controller (NIC) can write information reserving one or more cache lines in a first level of the ATC to a second level of the ATC. The NIC can receive a request for a direct memory access (DMA) to an untranslated address in memory of a host computing system. The NIC can determine that the untranslated address is not cached in the first level of the ATC. The NIC can identify a selected cache line in the first level of the ATC to evict using the request and the second level of the ATC. The NIC can receive a translated address for the untranslated address. The NIC can cache the untranslated address in the selected cache line. The NIC can perform the DMA using the translated address.

Abstract: Provided are techniques for destaging pinned retryable data in cache. A ranks scan structure is created with an indicator for each rank of multiple ranks that indicates whether pinned retryable data in a cache for that rank is destageable. A cache directory is partitioned into chunks, wherein each of the chunks includes one or more tracks from the cache. A number of tasks are determined for the scan of the cache. The number of tasks are executed to scan the cache to destage pinned retryable data that is indicated as ready to be destaged by the ranks scan structure, wherein each of the tasks selects an unprocessed chunk of the cache directory for processing until the chunks of the cache directory have been processed.

Abstract: Provided is a system and method for improving memory management in a database. In one example, the method may include receiving a request to store a data object within a database, determining a category type associated with the data object from among a plurality of category types based on an attribute of the data object, and storing the data object via a memory pool corresponding to the determined category from among a plurality of memory pools corresponding to the plurality of respective categories, where the storing comprises allocating a first category type of data object to a first memory pool locked to main memory and allocating a second category type of data object to a second memory pool that is swapped out to disk over time. The locked memory pool can ensure that more important data items remain available even when they are the least recently used.

Abstract: A buffer status metric representing a current amount of video packets in a video playback buffer (303) of a user device (300). A buffer status action is triggered based on the buffer status action and at least one of a mobility metric representing a mobility pattern of the user device (300) and a radio quality metric representing a signal strength of a radio channel carrying video packets towards the user device (300). The embodiments also relate to determination of a buffer control model (142) that can be used in order to control the video playback buffer based on the input metrics. The embodiments achieve a more efficient control of video playback buffers (303) and may reduce the risk of freezes during video playback.

Abstract: A system and method for efficiently storing data in a storage system. A data storage subsystem includes multiple data storage locations on multiple storage devices in addition to at least one mapping table. A data storage controller determines whether data to store in the storage subsystem has one or more patterns of data intermingled with non-pattern data within an allocated block. Rather than store the one or more pattern on the storage devices, the controller stores information in a header on the storage devices. The information includes at least an offset for the first instance of a pattern, a pattern length, and an identification of the pattern. The data may be reconstructed for a corresponding read request from the information stored in the header.

Abstract: A cache coherency protection system provides for data redundancy by sharing a cache coherence memory pool for protection purposes. The system works consistently across all communication protocols, yields improved data availability with potentially less memory waster and makes data availability faster in node/director failure scenarios. According to various embodiments, the cache coherency protection system may include a writer/requester director that receives a write request from host, a protection target director that is a partner of the writer/request director and a directory.

Abstract: Disclosed are embodiments for running an application on a local processor when the application is dependent on pages not locally present but contained in a remote host. The system is informed that the pages on which the application depends are locally present. While running, the application encounters a cache miss and a cache line satisfying the miss from the remote host is obtained and provided to the application. Alternatively, the page containing the cache line satisfying the miss is obtained and the portion of the page not including the cache line is stored locally while the cache line is provided to the application. The cache miss is discovered by monitoring coherence events on a coherence interconnect connected to the local processor. In some embodiments, the cache misses are tracked and provide a way to predict a set of pages to be pre-fetched in anticipation of the next cache misses.

Abstract: An infrequently used method is selected for eviction from a code cache repository by accessing a memory management data structure from an operating system, using the data structure to identify a first set of pages that are infrequently referenced relative to a second set of pages, determining whether or not a page of the first set of pages is part of a code cache repository and includes at least one method, in response to the page of the first set of pages being part of the code cache repository and including at least one method, flagging the at least one method as a candidate for eviction from the code cache repository, determining whether or not a code cache storage space limit has been reached for the code cache repository, and, in response to the storage space limit being reached, evicting the at least one flagged method from the code cache repository.