4667 results about "Magnetic disks" patented technology

Embodiments of the invention provide a write head for perpendicular recording in which the magnetic field intensity applied at a read element is reduced with larger write-field gradients and the distance between a write element and a read element is small. In one embodiment, a pair of magnetic bodies for shielding a read element are provided on the leading and trailing sides of the read element respectively and whichever one of the two shields is the longer in the length from the air bearing surface is shorter than the length up to the back contact position PBK at which a main pole and an auxiliary pole are connected.

The invention relates to the technical field of data storage, in particular to a dynamic data distribution storage method and system for an RAID (redundant array of independent disks), and the method comprises the following steps: based on the access frequency of data blocks in a disk, recording the access frequency and timestamp of each data block, carrying out heat rating on the data blocks, and determining thresholds of hot data and cold data. According to the method, the dynamic distribution of data storage is effectively optimized by analyzing the access frequency of the data blocks, and the performance and the economic efficiency of a storage system are remarkably improved by migrating the frequently accessed data to the cache disk and migrating the less accessed data to the low-cost storage; according to the method, unnecessary occupation of high-performance storage resources is reduced, the operation cost is also reduced, heat rating is performed on the data blocks, and the storage positions of the data blocks are correspondingly adjusted, so that the data processing speed is higher, meanwhile, the data security is improved, when a hard disk fails, data recovery is faster, and the service continuity is ensured.

Securely persisting transient data between virtual machine restarts or VM migrations involves terminating, by a first virtual machine (VM) during a shutdown process for the first VM, execution of user-space processes on the first VM, writing, by a first agent executing on the first VM, protected data from transient memory of the first VM to a virtual disk accessible by the first VM and shutting down the first VM. The process also involves initiating a startup process of a second VM, the second VM mounting the virtual disk; and executing, at the second VM and prior to execution of user-space processes, a second agent, the second agent being configured to: read the protected data from the virtual disk into transient memory of the second VM; and delete the protected data from the virtual disk.

The invention discloses a data read-write request processing method, electronic equipment, a storage medium and a program product, and relates to the technical field of redundant arrays of independent disks, and the data read-write request processing method comprises the following steps: if the category of a data read-write request is a random read request, if target data corresponding to the random read request does not exist in a target cache, executing the random read request; the random read request is directly issued to a corresponding redundant array of independent disks for processing, and pre-read processing is not carried out; if the category of the data read-write request is the random write request, the random write request is stored by using the random write cache and the random write backup cache, and when the proportion of the random read-write request is less than the preset proportion threshold, the random write request is issued to the corresponding redundant array of independent disks for processing, and the random write request does not need to be merged; and the number of the random write requests issued in a centralized manner is reduced, the technical problem that the performance of the storage system is reduced by a mode of processing the random read-write requests in related technologies is solved, and the technical effect of improving the performance of the storage system is achieved.

The invention discloses a data reconstruction method and device, a storage medium and a program product, and relates to the technical field of data storage, and the method comprises the following steps: reconstructing an initial redundant array of independent disks according to a hard disk state of the initial redundant array of independent disks to obtain a target redundant array of independent disks; after the reconstruction of the initial redundant array of independent disks is completed, establishing a mapping relation between the initial redundant array of independent disks and the target redundant array of independent disks, and storing the mapping relation based on metadata to obtain corresponding target metadata, and recovering the target redundant array of independent disks to the initial redundant array of independent disks according to the target metadata. When the disk fails or restores to normal for data reconstruction, the corresponding data can be directly written back according to the mapping condition between the blocks, additional check block calculation is not needed, and the data recovery speed is increased.

The invention provides a method for carrying out fault judgment based on a computer hard disk state index, and relates to the technical field of computers.The method comprises the steps that a periodic data collection mechanism is set, and multi-dimensional feature data including disk rotation instability, the data transmission rate, the read-write error rate and the like are obtained; carrying out normalization and de-noising processing on the data; then, data dimension reduction is achieved through an auto-encoder, key feature vectors are extracted, and then fault probability calculation is conducted through a support vector machine; on the basis, a threshold fine tuning model is constructed, and the fault probability threshold can be dynamically adjusted according to the relevance between the physical state and the operation state collected in real time and the change trend of the physical state and the operation state; the accuracy and timeliness of fault judgment are improved, the probability of false alarm is reduced through a dynamic adjustment mechanism, and the safety of data storage and the normal operation of the system are ensured.

A pair of waveform disks for use in a disk-pack turbine used, for example, in systems and methods in at least one embodiment for separating fluids including liquids and gases into subcomponents by passing the fluid into an expansion chamber and then through at least a portion of a waveform pattern present between at least two disks. The disks having waveform patterns on at least one side.

The invention provides a scheduling method and system for Direct IO intensive tasks under an NUMA architecture. The method comprises the following steps: creating a user mode scheduling program; a scheduling domain is established, the CPU utilization rate is initialized, and a bpf callback function is achieved; taking over the scheduling strategy of the schedext extensible scheduler; through interaction between the bpf map and a callback function, queuing and distribution of tasks are realized; marking Direct IO intensive tasks according to whether the IO quantity exceeds an IO quantity threshold value or not; the tasks are scheduled in the corresponding disk nodes; if the set utilization rate threshold value is exceeded, dynamically adjusting the task scheduling range; and determining whether the memory needs to be migrated to the currently accessed disk node or not according to a set threshold value. According to the method, the overhead caused by accessing the disk across the NUMA nodes is reduced, efficient scheduling is realized, and the task performance is improved.

The invention discloses a channel model dynamic association method based on a multi-source data index, and relates to channel engineering digital management. The channel model dynamic association method comprises the steps that component codes of channel engineering are constructed and used for identifying the same physical entity; establishing a three-layer index table structure according to the component codes; constructing a dynamic partition management mechanism of a three-level cache architecture, wherein the three-level cache architecture comprises a first-level cache stored in a memory, a second-level cache stored in a local disk and a third-level cache stored in a distributed manner; according to data access information in the three-layer index table structure, data migration is carried out among all levels of caches; according to the dynamic partition management mechanism, a data version management mechanism based on a timestamp concurrency control protocol is established, the data version management mechanism generates incremental data by comparing original data with updated data, and incremental updating or full updating is selected according to a threshold value. According to the method, the problem that data evolution tracking is difficult due to fragmentation of multi-source heterogeneous data and unbalanced access efficiency in navigation channel engineering is solved.

The invention discloses a large model training fault recovery method based on distributed memory management, and the method comprises the steps: obtaining a model state at a current moment when judging that a check point storage service needs to be carried out in a current training round, serializing the obtained model state into a storable format, and generating check point data; then transmitting the check point data to a host side memory from a device side memory in the node; after the transmission is finished, continuing to train the task, and storing asynchronous check points at the same time; the asynchronous check point storage comprises data distributed memory backup and data persistence to a disk; and when the training task goes wrong, task recovery is carried out through check point data. When the fault is recovered, the backup data is preferentially read from the memory, and the delay caused by the I / O bottleneck of the disk can be obviously reduced, so that the training pause or the efficiency reduction caused by the operation of storing or loading the check point data is reduced, the efficient utilization of the GPU or other computing resources is kept, and the continuity and the stability of large-scale model training are improved.

Embodiments of the present application relate to a method, a bridging device, a system and a medium of virtualization processing of a storage device. The method comprises: receiving an initial access request to a virtual disk sent by a virtual machine user; translating the virtual address corresponding to the virtual machine to a first physical address corresponding to a host based on a preconfigured address mapping relationship; translating the virtual access address to a second physical address corresponding to the storage device based on a preconfigured virtual partition mapping relationship; and generating a target access request based on the first physical address and the second physical address, and sending the target access request to the host, so as to cause the host to perform information interaction with the storage device based on the target access request.

The invention discloses a small file distributed high-concurrency aggregation storage method and system, and the method comprises the steps: collecting performance index data, and transmitting the data to a resource coordination server; calculating a queuing rate; determining a target server according to the queuing rate, the file reading concurrency and the disk residual space; determining a first-level directory hash value; selecting a first-level storage directory and a second-level storage directory according to the first-level directory hash value; creating a target aggregation file in the secondary storage directory, generating a 2-byte file format field according to the compression identifier or the non-compression identifier, and generating a 4-byte file length field according to the size of the to-be-written file; writing the file format field, the file length field and the to-be-written file into a target aggregation file in sequence, and recording the initial offset of writing; and generating a file name containing the storage position information, and returning the file name to the client. According to the invention, the efficiency and effect of small file storage are optimized.

The invention provides a magnetic sheet surface defect detection method based on an improved target detection model. The method comprises the following steps: constructing a defect detection model based on a YOLO11n basic network model; the backbone network comprises at least one convolution module, at least one C3k2Her module combined with an edge refining structure and a layered edge refining structure, a spatial pyramid pooling fast module and a C2PSA module combined with a cross-stage local structure and a pyramid extrusion attention mechanism; the neck network is constructed based on a path aggregation feature pyramid network structure and has a multi-layer feature pyramid structure for processing a low-layer feature map to be fused to a higher-layer feature map, and the neck network further comprises at least one C3k2FFCM module combined with a cross-stage local structure and a fused Fourier convolution mixer module; training the defect detection model; performing defect detection; and outputting the surface defect information. According to the invention, the detection precision and efficiency of the surface of the magnetic sheet, especially small defects under low contrast and complex backgrounds, are improved.

In one aspect, a highly scalable diffusion-couple apparatus includes a transfer chamber configured to load a wafer into a process chamber. The process chamber is configured to receive the wafer substrate from the transfer chamber. The process chamber comprises a chamber for growth of a diffusion material on the wafer. A heatable bottom disk includes a first heating mechanism. The heatable bottom disk is fixed and heatable to a specified temperature. The wafer is placed on the heatable bottom disk. A heatable top disk comprising a second heating mechanism. The heatable top disk is configured to move up and down along an x axis and an x prime axis to apply a mechanical pressure to the wafer on the heatable bottom disk. While the heatable top disk applies the mechanical pressure, a chamber pressure is maintained at a specified low value. The first heating mechanism and the second heating mechanism can be independently tuned to any value in the working range.

Embodiments of the present disclosure include a non-transitory computer readable medium that when executed by at least one processor cause the at least one processor to perform operations for dynamic cloud workload protection, the operations comprising: installing an agentless scanning system, the agentless scanning system being configured to scan a cloud server, the cloud server including a network and a memory; detecting, using a cloud provider application program interface (API), an installation of a new workload in the cloud server, the new workload including disks; scanning, using the agentless scanning system, the disks of the new workload; installing an agent on the new workload; monitoring, using the agent, the disks, the network, and the memory of the new workload; generating, using the agent, a notification when an interesting event occurs; scanning, using the agentless scanning system, the cloud server; and generating at least one command to perform one or more of a remediation or a policy update.

According to an embodiment, tracks on a magnetic disk each include a long-distance sector having a length in the circumferential direction covering two or more servo sectors. A controller executes an acquisition operation to acquire one or more evaluation amounts on the basis of a track pitch in each of the two or more servo sectors included in a portion adjacent to the long-distance sector. The controller executes a protection operation to protect data of an adjacent track in a case where a total value of the one or more evaluation amounts exceeds a first threshold value.

The embodiment of the invention provides a data read-write method, system and device, computing equipment and a storage medium, and the data read-write method comprises the following steps: receiving a data read-write request sent by a user mode through an asynchronous I / O layer, and storing the data read-write request into a request queue; under the condition that the asynchronous I / O layer determines that the data read-write request is of a straight-through type, the data read-write request is sent to a driving layer; and analyzing the data read-write request through the driving layer, and sending request information of the data read-write request to target processing end equipment for read-write processing. The asynchronous I / O layer receives the data read-write request sent by the user mode and sends the data read-write request to the driving layer after the data read-write request is determined to be of the straight-through type, context switching and data copying expenses are reduced, the processing efficiency of I / O operation is improved, the request is analyzed in the driving layer and sent to the processing end equipment for read-write processing, scheduling of I / O tasks is optimized, and the user experience is improved. The disk addressing time and I / O delay are reduced, and the overall performance and the system stability are improved.

The invention discloses a solid state disk controller, a solid state disk and a storage system, and relates to the technical field of storage, and the solid state disk controller comprises a communication endpoint unit which is used for determining a target working mode according to a control instruction sent by host equipment; if the target working mode is an exclusive-or accelerator mode, calling an exclusive-or computing engine, and processing a direct-connection disk management operation task issued by the host equipment; wherein the direct connection disk management operation task comprises a parity check operation task corresponding to disk equipment connected with the host equipment; the XOR calculation engine is used for carrying out operation processing on the direct connection disk management operation task according to the control of the communication endpoint unit; hardware acceleration of the RAID function of the host equipment is achieved through the solid state disk controller in the solid state disk connected with the host equipment, the burden of a host CPU is relieved, special hardware equipment does not need to be additionally arranged, and balance of high performance and low cost is achieved.

The invention discloses a caching method, device and equipment and a computer readable storage medium, and is applied to the field of caching, and the caching method comprises the following steps: screening data in a disk according to a screening condition, and writing the screened data into a read cache; the screening condition at least comprises one of data disk falling time, access frequency and read request data volume; calculating a cache proportion of the cache occupying the memory, and judging whether the cache proportion reaches a starting threshold value or not; and if so, triggering a deletion mechanism for dynamically adjusting the speed, and deleting the data in the read cache. According to the method, the data in the disk is screened according to the screening conditions, so that the amount of data written in the read cache can be reduced, and the cache efficiency can also be ensured; and when a deletion mechanism is triggered, the dynamic deletion speed is utilized, so that the influence of deletion on the read request can be reduced to the minimum extent under a relatively small cache pressure, and the read request cannot cause cache exhaustion to influence the overall performance under a relatively large pressure.

A base plate includes a support part, a bottom plate part, and a wall part. The wall part includes a first wall part and a second wall part above the first wall part. The first wall part includes a first wall surface and a support surface spreading from an upper end of the first wall surface to a radially outer side. The second wall part includes a second wall surface and a recess recessed to the radially outer side from the second wall surface. The recess is defined by a space enclosed by a recess wall surface located on the radially outer side of the second wall surface and extending upward from the support surface, a recess top surface connecting from an upper end of the recess wall surface to the second wall surface, and a recess support surface facing the recess top surface in an axial direction.

The invention discloses a garbage collection method and device, equipment and a storage medium, and relates to the technical field of electrical digital data processing.The method comprises the steps that when metadata of a cache pool is updated, internal object abstracts corresponding to the metadata are put into a garbage collection queue; according to the water level condition of the data pool and the disk utilization rate of the object storage device of the data pool, the single-round execution task number of the garbage collection queue is determined; according to the number of single-round execution tasks of the garbage collection queue and the garbage collection priorities of the internal objects, the internal objects are selected to execute garbage collection, so that effective data in the internal objects are aggregated into new internal objects, and the internal objects subjected to garbage collection are deleted, namely, the situation that the water level of the current data pool is too high, and the garbage collection efficiency is improved is avoided. And the overlarge disk utilization rate of the object storage equipment is avoided, and the two are balanced, so that the influence on other businesses is avoided as far as possible on the basis of ensuring the normal recovery of the garbage.

According to one embodiment, a magnetic disk device includes a disk, a write head, an adjustment unit, and a write processing unit. The adjustment unit adjusts each of a first adjustment value, a second adjustment value, and a third adjustment value. The write processing unit can select shingled magnetic recording and perform write processing based on the first to third adjustment values. When the number of the plurality of tracks of the band is t+1 and the number of unused sectors among the plurality of sectors of the band is e. The adjustment unit adjusts the first to third adjustment values to establish ext.

According to one embodiment, a disk device includes a magnetic disk, a magnetic head, a flexure, a piezoelectric element, a first bonding material, a second bonding material, and a protrusion. The flexure includes a first outer surface, a first pad, and a second pad. The first pad and the second pad are on the first outer surface. The piezoelectric element includes a second outer surface, a first electrode, and a second outer surface. The first electrode and the second electrode are on the second outer surface. The first bonding material, which is conductive, bonds the first pad and the first electrode. The second bonding material, which is conductive, bonds the second pad and the second electrode. The protrusion is provided on the flexure, is located at least partially between the first bonding material and the second bonding material, and protrudes from the first outer surface.

The invention relates to the field of computers, and provides a method and a device for pre-aggregating time series data. The method comprises the following steps: acquiring at least one time sequence data from a first device; the first device is a generation device of at least one time sequence data, or the first device is a node device of a time sequence database, and any time sequence data in the at least one time sequence data comprises an identifier; determining a pre-polymerization method according to the identifier of the at least one time sequence data, wherein the pre-polymerization method and the identifier of the at least one time sequence data have a preset binding relationship; processing the at least one time sequence data by using a pre-aggregation method to generate pre-aggregated data; and when the occupancy rate of the memory meets a trigger condition, writing the pre-aggregated data and the at least one time sequence data into a disk. According to the method, the time sequence data and the pre-aggregated data can be uniformly distributed, and the utilization rate of the storage space is improved.

According to one embodiment, a magnetic disk device includes a disk, a head configured to write data to the disk and read the data from the disk, and a controller configured to control write processing based on a first determination value corresponding to a first shift amount defined for the excessive number of times at which the first shift amount of the head in a radial direction of the disk exceeds a first threshold value causing a read error in a second track adjacent to a first track in the radial direction at a time of the write processing of the first track of the disk, and a second threshold value changing the write processing.

A system and method for inspecting encrypted disks for a cybersecurity object using a generic key is disclosed. The method includes: detecting an encrypted disk in a cloud computing environment, the cloud computing environment including a security policy service; authorizing a key policy on the security policy service for a default key of an inspector account, wherein the key policy is a policy authorized to decrypt the encrypted disk; generating a second encrypted disk based on the encrypted disk; inspecting the second encrypted disk for a cybersecurity object with the default key; and releasing a resource allocated to the second encrypted disk in response to completing the inspection.

The invention relates to the technical field of computers, and discloses a firmware upgrading method and system, computer equipment and a storage medium, the method is applied to a substrate management controller, and the method comprises the following steps: receiving a firmware upgrading request sent by a client when detecting that a target type of redundant array of independent disks is deployed on a first server; a firmware mirror image file in a first format included in the firmware upgrading request is converted into a data packet in a preset protocol format, the data packet carries a target endpoint identifier, and a preset protocol is a management component transmission protocol transmitted through a peripheral component interconnection fast link; and forwarding the data packet carrying the target endpoint identifier to the target redundant array of independent disks through the central processing unit, so that the target redundant array of independent disks performs firmware upgrading based on the data packet. According to the method and the device, the problems that when firmware upgrading is carried out on the redundant array card of the independent disk, a specific operating system is depended on, and currently executed services need to be interrupted can be solved.

The embodiments of the present application relate to the technical field of computers. Provided are a data processing method and apparatus, and an electronic device and a computer-readable storage medium. The method comprises: using a first hash function and a second hash function to encode, into a first hash value and a second hash value, each key in key value data to be stored; traversing first hash values and second hash values of all keys in said key value data, and determining parameter information of a perfect hash corresponding to said key value data, wherein the parameter information of the perfect hash comprises: an index of a slot corresponding to each key; and storing said key value data according to the index of the slot corresponding to each key. The solutions of the embodiments of the present application solve the problems of IO overheads of disks being high and the query efficiency being low during data query.

The invention provides an equipment fault prediction model determination method and device, equipment, a storage medium and a product. The method comprises the steps that a target neural network model used for equipment fault prediction and a disk data sample set are obtained, the disk data sample set comprises a plurality of disk data samples, and the disk data samples comprise disk state data of a disk and fault labels corresponding to the disk state data; the disk state data comprises feature data in multiple dimensions; inputting target feature data of each disk data sample into the target neural network model to obtain a contribution degree corresponding to the target feature data; constructing a contribution degree matrix according to the contribution degree corresponding to each piece of target feature data; and according to the contribution degree matrix, converting a target neural network model into a decision tree model taking the target feature data as nodes.