Formation imaging system, formation imaging method, device and medium
By introducing CXL2.0 memory pool technology into the reverse time migration stratigraphic imaging algorithm, the problem of huge memory requirements in the reverse time migration stratigraphic imaging algorithm is solved by dynamically allocating memory, realizing efficient management of storage resources within the cluster and reducing hardware resource consumption.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- CHINA TELECOM CLOUD TECH CO LTD
- Filing Date
- 2025-11-20
- Publication Date
- 2026-06-04
AI Technical Summary
Existing technologies in reverse time migration stratigraphic imaging algorithms have huge memory requirements, resulting in high hardware resource consumption and failing to effectively meet the storage needs within the cluster.
By employing CXL2.0 memory pool technology, data entry and retrieval during the forward and backward propagation simulation stages of the reverse time migration stratigraphic imaging algorithm are achieved through dynamic memory allocation, thereby reducing hardware resource consumption.
The CXL2.0 memory pool technology meets the dynamic memory requirements of the inverse time-migration stratigraphic imaging algorithm within the cluster, effectively manages the shared memory pool within the cluster, and reduces the consumption of hardware resources.
Smart Images
Figure CN2025136471_04062026_PF_FP_ABST
Abstract
Description
A stratigraphic imaging system, a stratigraphic imaging method, an apparatus, and a medium
[0001] Cross-reference to related applications
[0002] This application claims priority to Chinese Patent Application No. 202411706809.6, filed on November 26, 2024, entitled "A Stratigraphic Imaging System, a Method, Apparatus and Medium for Stratigraphic Imaging", the entire contents of which are incorporated herein by reference. Technical Field
[0003] This application relates to the technical field of stratigraphic imaging, and in particular to a stratigraphic imaging system, a stratigraphic imaging method, an electronic device, and a computer-readable storage medium. Background Technology
[0004] In the field of oil and gas exploration, Reverse Time Migration (RTM) technology has become the mainstream deep formation imaging technology for global oil companies. RTM is used to depict the Earth's subsurface at high resolution. The algorithm is based on a two-way wave equation and imaging conditions.
[0005] During the algorithm's implementation, the source wavefield needs to be constantly exchanged between main memory and the high-speed memory of the GPU (Graphics Processing Unit) to update the forward and reverse wavefields spatially and temporally, and to calculate imaging conditions. This process places enormous demands on memory. How to meet the memory requirements of the RTM algorithm within the cluster has become one of the most pressing problems to be solved. Summary of the Invention
[0006] In view of the above problems, a stratigraphic imaging system, a stratigraphic imaging method, an electronic device, and a computer-readable storage medium are proposed to overcome or at least partially solve the above problems, comprising:
[0007] A stratigraphic imaging system is provided, which performs stratigraphic imaging based on a reverse time migration stratigraphic imaging algorithm; the system includes: at least one set of nodes, and a CXL2.0 (Compute Express Link 2.0) memory pool communicating with at least one set of nodes;
[0008] At least one set of nodes is used to input the computation process data into the CXL2.0 memory pool during the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm; and to read the computation process data from the CXL2.0 memory pool during the backward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm.
[0009] The CXL2.0 memory pool is used to dynamically allocate memory for at least one group of nodes.
[0010] Optionally, for any set of nodes, including master nodes and slave nodes;
[0011] The master node is used to reserve CXL (Compute Express Link) memory pool resources in the CXL2.0 memory pool; when the reservation is successful, the master node and slave nodes will enter the computation process data into the CXL2.0 memory pool.
[0012] Optionally, any node includes a graphics processing unit and a central processing unit, wherein the graphics processing unit reserves at least one first data slot and the central processing unit reserves at least one second data slot;
[0013] The master node is used to transfer computation process data from the first data slot to the second data slot according to a preset time step during the forward propagation simulation phase; and during the backward propagation simulation phase, when the first data slot is idle, to push computation process data from the second data slot into the first data slot.
[0014] Optionally, at least one set of nodes is also used to destroy the corresponding computation process data in the CXL2.0 memory pool after reading the computation process data from the CXL2.0 memory pool.
[0015] Optionally, at least one group of nodes includes at least a first group of nodes and a second group of nodes;
[0016] The first and second node groups alternately call the CXL2.0 memory pool.
[0017] Optionally, the first node group performs the step of recording the calculation process data into the CXL2.0 memory pool between the first time point and the second time point; and performs the step of reading the calculation process data from the CXL2.0 memory pool between the second time point and the third time point.
[0018] The second node group executes the step of recording the calculation process data into the CXL2.0 memory pool between the second and third time points.
[0019] This application also provides a method for stratigraphic imaging, applied to the above-mentioned system, the system including: at least one set of nodes, and a CXL2.0 memory pool communicating with the at least one set of nodes; the method includes:
[0020] During the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm, the data from the first calculation process is entered into the CXL2.0 memory pool;
[0021] In the backpropagation simulation stage of the reverse time migration stratigraphic imaging algorithm, the first calculation process data is read from the CXL2.0 memory pool, and the target stratigraphic image is generated based on the first calculation process data.
[0022] Optionally, at least one set of nodes includes at least a first node group and a second node group; during the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm, the data from the first calculation process is entered into the CXL2.0 memory pool, including:
[0023] Between the first time point and the second time point, the data from the first calculation process is entered into the CXL2.0 memory pool;
[0024] During the backpropagation simulation phase of the reverse time migration stratigraphic imaging algorithm, the first computational process data is read from the CXL2.0 memory pool, including:
[0025] Between the second and third time points, the first calculation process data is read from the CXL2.0 memory pool;
[0026] The second node group is used to input the second calculation process data into the CXL2.0 memory pool between the second time point and the third time point.
[0027] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the above-described method for stratigraphic imaging.
[0028] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for stratigraphic imaging.
[0029] The embodiments of this application have the following advantages:
[0030] In this embodiment, the system performs stratigraphic imaging based on a reverse time migration (RTM) stratigraphic imaging algorithm. The system includes at least one set of nodes and a CXL2.0 memory pool that communicates with the at least one set of nodes. The at least one set of nodes is used to input computational process data into the CXL2.0 memory pool during the forward propagation simulation phase of the RTM stratigraphic imaging algorithm and to read computational process data from the CXL2.0 memory pool during the backward propagation simulation phase of the RTM stratigraphic imaging algorithm. The CXL2.0 memory pool is used to dynamically allocate memory to the at least one set of nodes.
[0031] The embodiments of this application can meet the dynamic memory requirements of the inverse time migration stratigraphic imaging (RTM) algorithm within the cluster, and achieve effective management of the public memory pool within the cluster, thereby reducing the consumption of hardware resources. Attached Figure Description
[0032] To more clearly illustrate the technical solution of this application, the drawings used in the description of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0033] Figure 1a is a schematic diagram of a data scheduling process in the prior art;
[0034] Figure 1b is a flowchart of a data scheduling process in the prior art;
[0035] Figure 2 is a schematic diagram of the structure of a formation imaging system according to an embodiment of this application;
[0036] Figure 3 is a flowchart of the steps of a stratigraphic imaging method according to an embodiment of this application;
[0037] Figure 4 is a flowchart of another method for stratigraphic imaging according to an embodiment of this application;
[0038] Figure 5 is a schematic diagram of a data scheduling process according to an embodiment of this application;
[0039] Figure 6 is a schematic diagram of another data scheduling process according to an embodiment of this application;
[0040] Figure 7 is a schematic diagram of an interleaved scheduling according to an embodiment of this application;
[0041] Figure 8 is a flowchart of a data scheduling process according to an embodiment of this application. Detailed Implementation
[0042] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0043] RTM is considered a time-consuming and data-intensive technique. Currently, solving the control wave equations of RTM using numerical methods is computationally very complex, and discretization using the finite difference method also leads to higher computational complexity for more accurate simulations.
[0044] Fortunately, RTM-based imaging algorithms are awkwardly parallel, with each seismic shot simulation being completely independent. Typically, only one seismic shot needs to be calculated by coordinating multiple GPU servers.
[0045] Specifically, as shown in Figure 1a, users (Users1) can use HPC cabinet 1 in HPC (High-Performance Computing) cluster 1 to complete the task of ground imaging through front-end (Frontal1). HPC cabinet 1 can include multiple nodes.
[0046] However, several problems remain to be solved with the existing technical solutions:
[0047] 1. Because GPUs are much faster than CPUs (Central Processing Units), node memory cannot be expanded proportionally due to procurement costs and host slot limitations.
[0048] 2. During forward propagation wavefield calculation, the memory requirement of the RTM algorithm increases linearly, peaks before backpropagation begins, and then decreases linearly with backpropagation. The actual limitation of the algorithm is the maximum memory value of the node.
[0049] To address the first problem, a common technical solution is to use lossy compression algorithms (such as ZFP (a floating-point data compression library) or WPC (Wall Power Consumption) on the source wave field. These algorithms can achieve a compression ratio of nearly 10 times, significantly alleviating storage requirements. The specific algorithm structure can be seen in Figure 1b.
[0050] First, the RTM (Reverse Time Migration) shot record can be initialized; specifically, a reverse time migration (RTM) calculation task can be started first, and the relevant data and parameters can be initialized.
[0051] Then, the data can be moved to the GPU (Graphics Processing Unit); specifically, the data required for computation can be transferred from the host memory (CPU) to the GPU memory so that computation can be performed on the GPU.
[0052] Next, the forward wave field can be propagated; specifically, the propagation calculation of the forward wave field can be performed on the GPU to simulate the propagation process of seismic waves in the underground medium.
[0053] When propagating a forward wave field, the forward wave field can be compressed; specifically, the calculated forward wave field data can be compressed to reduce storage space and transmission time.
[0054] Then, the forward wave field can be moved to the host; specifically, the compressed forward wave field data can be transferred from GPU memory back to host memory.
[0055] After completing a seismic test, it can be checked whether the time has reached Tmax (Maximum testing time) from Tmin (Minimum testing time). If not, the step of propagating the forward wave field is re-executed. If Tmax is reached, the forward wave field can be moved to the GPU. Specifically, the forward wave field data can be transferred to the GPU.
[0056] Decompress the forward wave field; specifically, the forward wave field data transmitted to the GPU can be decompressed to restore the original data format.
[0057] Propagate the reverse wave field; perform reverse wave field propagation calculations on the GPU to simulate the process of seismic waves returning from the receiver to the source.
[0058] Calculate imaging conditions; based on the propagation results of the forward and reverse wave fields, calculate imaging conditions to generate images of underground structures.
[0059] Next, it can be checked whether the time has reached Tmin from Tmax; if not, the step of moving the positive wave field to the host is re-executed; if Tmin is reached, the host image can be updated and the result saved; specifically, the calculated imaging result can be updated to the image data in the host memory, and the final result can be saved to the storage device.
[0060] Even so, because the RTM algorithm needs to store the forward propagation wavefield, this scheme still cannot fundamentally reduce the RTM algorithm's dependence on the total memory of the nodes. In fact, the size of the node memory directly determines the number of GPU servers required per shot, thus indirectly affecting parallel efficiency.
[0061] To completely solve the above problems, this application embodiment introduces memory pooling technology, which successfully meets the dynamic memory requirements of the inverse time migration stratigraphy (RTM) algorithm within the cluster, and realizes effective management of the public memory pool within the cluster, thereby reducing the consumption of hardware resources.
[0062] Specifically, based on the above theory, this application provides a stratigraphic imaging system, as shown in Figure 2. This stratigraphic imaging system 10 performs stratigraphic imaging based on a reverse time migration stratigraphic imaging algorithm; the stratigraphic imaging system 10 may include:
[0063] At least one set of nodes 210, and a CXL2.0 memory pool 220 communicating with at least one set of nodes 210.
[0064] At least one set of nodes 210 is used to input the calculation process data into the CXL2.0 memory pool 220 during the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm; and to read the calculation process data from the CXL2.0 memory pool 220 during the backward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm.
[0065] CXL2.0 memory pool 220 is used to dynamically allocate memory for at least one set of nodes 210.
[0066] In practical applications, Reverse Time Migration (RTM) is an advanced imaging technique for generating high-resolution images of subsurface structures. RTM reconstructs images of subsurface structures using seismic data by simulating the propagation of seismic waves in the subsurface medium. The following are the basic steps and principles of the reverse time migration stratigraphic imaging algorithm.
[0067] The basic principle of reverse time migration is to reconstruct images of underground structures using seismic data by simulating the forward and backward propagation processes of seismic waves. The specific steps are as follows:
[0068] Forward propagation simulation: Simulates the process of seismic waves propagating from a seismic source through the subsurface medium.
[0069] Data acquisition: Recording data received by detectors during the propagation of seismic waves underground.
[0070] Backpropagation simulation: Using received seismic data, the process of seismic waves propagating backward from the detector location back into the ground is simulated.
[0071] Imaging: Images of underground structures are generated by cross-correlating the forward and backward propagating wavefields.
[0072] The steps of reverse time migration include data preparation, forward propagation simulation, back propagation simulation, and imaging; among which:
[0073] 1. Data preparation includes:
[0074] Seismic data: including the excitation signal from the seismic gun and the seismic wave data received by the detector.
[0075] Velocity model: A model that describes the velocity distribution in underground media.
[0076] Other parameters include time step and spatial grid size.
[0077] 2. Forward propagation simulation includes:
[0078] In the forward propagation simulation phase, the algorithm simulates the process of seismic waves propagating from the seismic gun into the underground medium.
[0079] Generating seismic waves: Seismic guns generate seismic waves, which propagate underground.
[0080] Wave propagation: When seismic waves encounter interfaces between different media underground, they are reflected, refracted, and scattered.
[0081] Data acquisition: The reflected waves are received by seismic detectors on the ground or in wells, and the arrival time and amplitude of the seismic waves are recorded.
[0082] 3. Backpropagation simulation includes:
[0083] In the backpropagation simulation phase, the algorithm simulates the process of seismic waves propagating backward from the received seismic data back underground.
[0084] Backward propagation: Using the received seismic data, the seismic waves are simulated to propagate backward from the detector location back underground.
[0085] Wave field reconstruction: Reconstructing the wave field of the underground structure through reverse propagation.
[0086] 4. Imaging includes:
[0087] During the imaging phase, the algorithm cross-correlates the forward and backward propagating wavefields to generate images of the underground structures.
[0088] Cross-correlation: Cross-correlation of the forward and backward propagating wave fields generates an image of the underground structure.
[0089] Image enhancement: Enhances and filters the generated image to improve its resolution and clarity.
[0090] The stratigraphic imaging system 20 may include at least one set of nodes 210, and a CXL (Compute Express Link) 2.0 memory pool communicating with the at least one set of nodes 210. Wherein:
[0091] In at least one group of nodes 210, any group of nodes 210 may include multiple nodes 210, and these multiple nodes 210 can communicate with the CXL2.0 memory pool 220 respectively.
[0092] As a key new technology in the development of information technology, CXL is becoming increasingly mature. By supporting high-performance, low-latency connectivity, CXL enables flexible access to components such as processors, accelerators, and memory, improving overall system performance and providing advanced error handling and reliability features.
[0093] It has received widespread support from numerous technology companies in terms of standard setting and promotion, and has gradually become the mainstream unified standard in the industry.
[0094] Furthermore, the continuous improvement of CXL at the software level, including enhancements to drivers, tools, and libraries, provides developers with a more convenient way to leverage its performance advantages, enabling the development of more efficient applications and solutions.
[0095] The CXL 2.0 standard provides memory pooling capabilities, which bind multiple CXL memory blocks on the server platform together to form a shared memory pool, providing significant cost benefits and resource utilization for each group of nodes 210.
[0096] In practical applications, each of the nodes 210 in at least one group of nodes 210 can input the calculation process data into the CXL2.0 memory pool 220 during the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm. The calculation process data can refer to the intermediate process data generated during the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm, such as forward wave field, reverse wave field, etc. This application embodiment does not limit this.
[0097] On the other hand, each group of nodes 210 can also read previously entered calculation process data from the CXL2.0 memory pool 220 for data processing during the backpropagation simulation stage of the reverse time migration stratigraphic imaging algorithm. This application embodiment does not limit this.
[0098] In practical applications, the CXL2.0 memory pool 220 can be used to dynamically allocate memory for at least one group of nodes 210. The CXL2.0 memory pool 220 can meet the memory requirements of the reverse time migration stratigraphic imaging algorithm of each node 210 and effectively manage the shared memory pool within the cluster, thereby reducing the consumption of hardware resources by the reverse time migration stratigraphic imaging algorithm.
[0099] In this embodiment, the system performs stratigraphic imaging based on a reverse time migration stratigraphic imaging algorithm. The system includes at least one set of nodes 210 and a CXL2.0 memory pool 220 communicating with the at least one set of nodes 210. The at least one set of nodes 210 is used to input calculation process data into the CXL2.0 memory pool 220 during the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm, and to read calculation process data from the CXL2.0 memory pool 220 during the backward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm. The CXL2.0 memory pool 220 is used to dynamically allocate memory to the at least one set of nodes 210.
[0100] The embodiments of this application can meet the dynamic memory requirements of the inverse time migration stratigraphic imaging (RTM) algorithm within the cluster, and achieve effective management of the public memory pool within the cluster, thereby reducing the consumption of hardware resources.
[0101] In one embodiment of this application, any group of nodes 210 includes a master node and slave nodes;
[0102] The master node is used to reserve CXL memory pool resources in CXL2.0 memory pool 220; when the reservation is successful, the master node and slave nodes will enter the calculation process data into CXL2.0 memory pool 220.
[0103] In some feasible embodiments, any set of nodes 210 may include a master node and other nodes 210 (hereinafter referred to as slave nodes).
[0104] For any group of nodes 210, the master node can communicate with other groups of master nodes and initialize the CXL2.0 memory pool 220.
[0105] When preparing to process the next seismic shot, a forward propagation simulation initialization can be performed first; during the forward propagation simulation phase, the master node can reserve resources first.
[0106] Specifically, the master node can first reserve CXL memory pool resources in CXL2.0 memory pool 220. If the reservation fails, it will continue to wait.
[0107] If the reservation is successful, the master node and the slave nodes in the same group to which the master node belongs can enter the calculation process data into the CXL2.0 memory pool 220.
[0108] In one embodiment of this application, any node 210 includes a graphics processing unit and a central processing unit. The graphics processing unit has at least one first data slot reserved, and the central processing unit has at least one second data slot reserved.
[0109] The master node is used to transfer computation process data from the first data slot to the second data slot according to a preset time step during the forward propagation simulation phase; and during the backward propagation simulation phase, when the first data slot is idle, to push computation process data from the second data slot into the first data slot.
[0110] In some feasible embodiments, for any node 210, the node 210 may include a graphics processing unit and a central processing unit; wherein:
[0111] At least one source data slot, i.e., the first data slot, can be reserved in the graphics processing unit; and at least one source data slot, i.e., the second data slot, can be reserved in the central processing unit.
[0112] For any group of nodes 210, the master node can transmit the calculation process data from the first data slot of the graphics processing unit to the second data slot of the central processing unit according to the preset time step during the forward propagation simulation phase.
[0113] Alternatively, for the master node, during the backpropagation simulation phase, when the first data slot of the graphics processing unit is idle, the computation process data in the second data slot of the central processing unit can be pushed from the second data slot of the graphics processing unit into the first data slot. This application embodiment does not limit this.
[0114] For example, the same steps can be performed on a slave node; specifically:
[0115] During the forward propagation simulation phase, the nodes can transmit computation process data from the first data slot of the graphics processing unit to the second data slot of the central processing unit according to a preset time step.
[0116] Alternatively, for slave nodes, during the backpropagation simulation phase, when the first data slot of the graphics processing unit is idle, the computation process data in the second data slot of the central processing unit can be pushed from the second data slot of the graphics processing unit into the first data slot. This application embodiment does not limit this.
[0117] In one embodiment of this application, at least one set of nodes 210 is further configured to destroy the corresponding computation process data in the CXL2.0 memory pool 220 after reading the computation process data from the CXL2.0 memory pool 220.
[0118] In some feasible embodiments, for any group of nodes 210, after reading the computation process data from the CXL2.0 memory pool 220, the corresponding computation process data in the CXL2.0 memory pool 220 can be destroyed so that nodes 210 in other groups can input data.
[0119] For example, for the master node, after reading the computation process data from the CXL2.0 memory pool 220, the corresponding computation process data in the CXL2.0 memory pool 220 can be destroyed.
[0120] For slave nodes, after reading the computation process data from the CXL2.0 memory pool 220, the corresponding computation process data in the CXL2.0 memory pool 220 can also be destroyed. This application embodiment does not limit this.
[0121] In one embodiment of this application, at least one group of nodes 210 includes at least a first node group and a second node group;
[0122] The first and second node groups alternately call CXL2.0 memory pool 220.
[0123] In some feasible embodiments, when the total amount in the CXL2.0 memory pool 220 is much greater than the total memory required for single-gun computing of all nodes 210 groups, the RTM single-gun resource (node 210 groups) scheduling strategy reverts to the normal mode, that is, it is called when there is demand.
[0124] When the total amount in the CXL2.0 memory pool 220 is less than the total memory required for single-shot calculation of all nodes 210, the resource usage needs to be controlled. Specifically, since the memory requirements of the forward propagation simulation stage and the backward propagation simulation stage are completely different, this embodiment of the application can control the first node group and the second node group to call the CXL2.0 memory pool 220 alternately.
[0125] For example, while the first node group is performing a forward propagation simulation phase, recording the computational process data into the CXL2.0 memory pool 220, the second node group can perform a backward propagation simulation phase to read data from the CXL2.0 memory pool 220; conversely, while the first node group is performing a backward propagation simulation phase, reading data from the CXL2.0 memory pool 220, the second node group can perform a forward propagation simulation phase to record the computational process data into the CXL2.0 memory pool 220. This interleaved usage maximizes resource utilization.
[0126] In one embodiment of this application, the first node group performs the step of recording the calculation process data into the CXL2.0 memory pool 220 between the first time point and the second time point; and performs the step of reading the calculation process data from the CXL2.0 memory pool 220 between the second time point and the third time point.
[0127] The second node group, between the second and third time points, performs the step of recording the calculation process data into the CXL2.0 memory pool 220.
[0128] In some feasible embodiments, the steps of interleaving calls to CXL2.0 memory pool 220 can be implemented in the following way:
[0129] Specifically, between the first time point and the second time point, each node 210 in the first node group can perform the step of recording the calculation process data into the CXL2.0 memory pool 220; between the second time point and the third time point, each node 210 in the first node group can perform the step of reading the calculation process data from the CXL2.0 memory pool 220.
[0130] Additionally, the second node group performs the step of recording the calculation process data into the CXL2.0 memory pool 220 between the second and third time points.
[0131] The order of the first time point, the second time point, and the third time point can be: first time point → second time point → third time point, but this application embodiment does not limit this.
[0132] Based on the formation imaging system mentioned in the above embodiments, this application also provides a formation imaging method. Referring to FIG3, FIG3 shows a flowchart of the steps of a formation imaging method according to an embodiment of this application.
[0133] As shown in Figure 3, the method for formation imaging may include the following steps:
[0134] Step 301: In the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm, the data of the first calculation process is entered into the CXL2.0 memory pool.
[0135] In practical applications, at least one set of nodes can input the first calculation process data into the CXL2.0 memory pool during the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm.
[0136] Step 302: In the backpropagation simulation stage of the reverse time migration stratigraphic imaging algorithm, read the first calculation process data from the CXL2.0 memory pool and generate the target stratigraphic image based on the first calculation process data.
[0137] On the other hand, this group of nodes can also read the previously entered first calculation process data from the CXL2.0 memory pool for data processing during the backpropagation simulation stage of the reverse time migration stratigraphic imaging algorithm. This application does not impose limitations on this aspect.
[0138] For example, after obtaining the first calculation process data, the nodes in this group can generate a stratigraphic image based on the first calculation process data, thereby obtaining the target stratigraphic image.
[0139] In this embodiment of the application, during the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm, the first calculation process data is entered into the CXL2.0 memory pool; during the backward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm, the first calculation process data is read from the CXL2.0 memory pool, and the target stratigraphic image is generated based on the first calculation process data.
[0140] The embodiments of this application can meet the dynamic memory requirements of the inverse time migration stratigraphic imaging (RTM) algorithm within the cluster, and achieve effective management of the public memory pool within the cluster, thereby reducing the consumption of hardware resources.
[0141] Referring to Figure 4, a flowchart of another method for stratigraphic imaging according to an embodiment of this application is shown, which may include the following steps:
[0142] Step 401: At least one group of nodes includes at least a first node group and a second node group; between the first time point and the second time point, the data of the first calculation process is entered into the CXL2.0 memory pool.
[0143] In some feasible implementations, the steps of interleaving calls to the CXL2.0 memory pool can be implemented in the following way:
[0144] Specifically, each node in the first node group can perform the step of recording the calculation process data into the CXL2.0 memory pool between the first time point and the second time point.
[0145] Step 402: Between the second time point and the third time point, read the first calculation process data from the CXL2.0 memory pool; the second node group is used to record the second calculation process data into the CXL2.0 memory pool between the second time point and the third time point.
[0146] Between the second and third time points, each node in the first node group can execute the step of reading computation process data from the CXL2.0 memory pool.
[0147] Additionally, the second node group performs the step of recording the calculation process data into the CXL2.0 memory pool between the second and third time points.
[0148] The order of the first time point, the second time point, and the third time point can be: first time point → second time point → third time point, but this application embodiment does not limit this.
[0149] For example, as shown in Figure 5, a user (Users2) can use the HPC cabinet2 in HPC cluster2 through the frontend (Frontal2) to complete the task of ground imaging. HPC cabinet2 can include multiple nodes. These nodes can interact with the nodes of the CXL2.0 memory pool to use the memory resources of the CXL2.0 memory pool.
[0150] As shown in Figure 6, a read / write control valve can be set for each group of nodes. For each group of seismic guns, the GPU memory can interact with the CPU memory, and the CPU memory can be controlled by the control valve to interact with the CXL2.0 memory pool. Different groups of control valves can communicate with each other via MPI (Message Passing Interface) to control the resource scheduling of the CXL2.0 memory pool.
[0151] For different groups of nodes, the resources of the CXL2.0 memory pool (CXL2.0 nodes) can be called in an interleaved manner; specifically, as shown in Figure 7, the forward propagation calculation and backward propagation calculation of multiple nodes can be staggered in the time dimension.
[0152] As shown in Figure 8, firstly, the RTM (Reverse Time Migration) shot record can be initialized; specifically, a reverse time migration (RTM) calculation task can be started first, and the relevant data and parameters can be initialized.
[0153] Then, the data can be moved to the GPU; specifically, the data required for computation can be transferred from host memory (CPU) to GPU memory so that computation can be performed on the GPU.
[0154] Next, the forward wave field can be propagated; specifically, the propagation calculation of the forward wave field can be performed on the GPU to simulate the propagation process of seismic waves in the underground medium.
[0155] When propagating a forward wave field, the forward wave field can be compressed; specifically, the calculated forward wave field data can be compressed to reduce storage space and transmission time.
[0156] Then, the forward wave field can be moved to the host; specifically, the compressed forward wave field data can be transferred from GPU memory back to host memory and recorded into the CXL2.0 memory pool.
[0157] After completing a seismic shot, it can be checked whether the time has reached Tmax from Tmin; if not, the step of propagating the forward wave field is re-executed; if Tmax is reached, the forward wave field can be moved to the GPU; specifically, the forward wave field data can be read from the CXL2.0 memory pool and the forward wave field data can be transferred to the GPU.
[0158] Decompress the forward wave field; specifically, the forward wave field data transmitted to the GPU can be decompressed to restore the original data format.
[0159] Propagate the reverse wave field; perform reverse wave field propagation calculations on the GPU to simulate the process of seismic waves returning from the receiver to the source.
[0160] Calculate imaging conditions; based on the propagation results of the forward and reverse wave fields, calculate imaging conditions to generate images of underground structures.
[0161] Next, it can be checked whether the time has reached Tmin from Tmax; if not, the step of moving the positive wave field to the host is re-executed; if Tmin is reached, the host image can be updated and the result saved; specifically, the calculated imaging result can be updated to the image data in the host memory, and the final result can be saved to the storage device.
[0162] In this embodiment, at least one group of nodes includes at least a first node group and a second node group. Between a first time point and a second time point, first calculation process data is entered into the CXL2.0 memory pool. Between a second time point and a third time point, first calculation process data is read from the CXL2.0 memory pool. The second node group is used to enter second calculation process data into the CXL2.0 memory pool between the second time point and the third time point. This embodiment improves resource utilization and allows for the processing of more complex and accurate simulations with the same total memory allocation.
[0163] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.
[0164] This application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the above-described method for stratigraphic imaging.
[0165] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for stratigraphic imaging.
[0166] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.
[0167] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0168] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0169] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.
[0170] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.
[0171] These computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable terminal equipment, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.
[0172] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.
[0173] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the element.
[0174] The above provides a detailed description of a formation imaging system, a formation imaging method, an electronic device, and a computer-readable storage medium. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A formation imaging system characterized by, The system performs stratigraphic imaging based on a reverse time migration stratigraphic imaging algorithm; the system includes: at least one set of nodes, and a CXL2.0 memory pool communicating with the at least one set of nodes; The at least one set of nodes is used to input the calculation process data into the CXL2.0 memory pool during the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm; and to read the calculation process data from the CXL2.0 memory pool during the backward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm. The CXL2.0 memory pool is used to dynamically allocate memory to the at least one group of nodes.
2. The system according to claim 1, characterized in that, For any set of nodes, including master nodes and slave nodes; The master node is used to reserve CXL memory pool resources in the CXL2.0 memory pool; when the reservation is successful, the master node and the slave node record the calculation process data into the CXL2.0 memory pool.
3. The system according to claim 2, characterized in that, Any node includes a graphics processing unit and a central processing unit. The graphics processing unit has at least one first data slot reserved, and the central processing unit has at least one second data slot reserved. The master node is used to transmit the computation process data from the first data slot to the second data slot according to a preset time step during the forward propagation simulation phase; and during the backward propagation simulation phase, when the first data slot is idle, to push the computation process data in the second data slot from the second data slot into the first data slot.
4. The system according to claim 1, characterized in that, The at least one set of nodes is further configured to destroy the corresponding computation process data in the CXL2.0 memory pool after reading the computation process data from the CXL2.0 memory pool.
5. The system according to claim 1, characterized in that, The at least one group of nodes includes at least a first node group and a second node group; The first node group and the second node group call the CXL2.0 memory pool alternately.
6. The system according to claim 5, characterized in that, The first node group performs the step of recording the calculation process data into the CXL2.0 memory pool between the first time point and the second time point; and performs the step of reading the calculation process data from the CXL2.0 memory pool between the second time point and the third time point. The second node group performs the step of recording the calculation process data into the CXL2.0 memory pool between the second time point and the third time point.
7. A method for stratigraphic imaging, characterized in that, Applied to a system as described in any one of claims 1-6, the system comprising: at least one set of nodes, and a CXL2.0 memory pool communicating with the at least one set of nodes; the method comprising: During the forward propagation simulation phase of the reverse time migration stratigraphic imaging algorithm, the data from the first calculation process is entered into the CXL2.0 memory pool; During the backpropagation simulation stage of the inverse time migration stratigraphic imaging algorithm, the first calculation process data is read from the CXL2.0 memory pool, and the target stratigraphic image is generated based on the first calculation process data.
8. The method according to claim 7, characterized in that, The at least one set of nodes includes at least a first node group and a second node group; the step of recording the first calculation process data into the CXL2.0 memory pool during the forward propagation simulation stage of the reverse time migration stratigraphic imaging algorithm includes: Between the first time point and the second time point, the data from the first calculation process is entered into the CXL2.0 memory pool; The step of reading the first calculation process data from the CXL2.0 memory pool during the backpropagation simulation phase of the reverse time migration stratigraphic imaging algorithm includes: Between the second and third time points, the first calculation process data is read from the CXL2.0 memory pool; The second node group is used to input the second calculation process data into the CXL2.0 memory pool between the second time point and the third time point.
9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method of stratigraphic imaging as described in any one of claims 7 to 8.
10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the method of stratigraphic imaging as described in any one of claims 7 to 8.