CN104570063A - Parallel extraction method of seismic channel set of seismic data - Google Patents

Parallel extraction method of seismic channel set of seismic data Download PDF

Info

Publication number
CN104570063A
CN104570063A CN201510073212.7A CN201510073212A CN104570063A CN 104570063 A CN104570063 A CN 104570063A CN 201510073212 A CN201510073212 A CN 201510073212A CN 104570063 A CN104570063 A CN 104570063A
Authority
CN
China
Prior art keywords
seismic
data
seismic channel
channel set
back end
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510073212.7A
Other languages
Chinese (zh)
Other versions
CN104570063B (en
Inventor
武杰
张捷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei keen Storage Technology Co., Ltd.
Original Assignee
Anhui Jituo Electronic Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Anhui Jituo Electronic Technology Co Ltd filed Critical Anhui Jituo Electronic Technology Co Ltd
Priority to CN201510073212.7A priority Critical patent/CN104570063B/en
Publication of CN104570063A publication Critical patent/CN104570063A/en
Application granted granted Critical
Publication of CN104570063B publication Critical patent/CN104570063B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention relates to a parallel extraction method of a seismic channel set of seismic data. The method comprises the following steps: writing a seismic data file into a disc by virtue of a homogenized storage technology; after a client sends a request command of reading a seismic channel set, sending the seismic channel set command to each data node; parsing the seismic channel set command by the data node and converting into a queue for reading a seismic request, then reading the seismic channel data in the local disc according to the queue, and finally returning the seismic data to the client; and receiving the seismic data by the client and sequencing the seismic data according to the channel set and finally returning the data to seismic data processing software. The method provided by the invention solves the problem that the efficiency is low conventionally based on a disc array system as the seismic channel set is extracted parallelly, so that the I/O efficiency of the seismic data is greatly improved, and the expandability is strong; the times of executing non-continuous seismic channel set instructions at the data nodes tend to be uniform, so that a theoretical expected extraction efficiency of a memory system can be realized.

Description

A kind of seismic channel set paralleling abstracting method of geological data
Technical field
The present invention relates to mass seismic data memory technology and equipment in a kind of earth exploration, particularly relate to a kind of seismic channel set paralleling abstracting method of geological data.
Background technology
Along with the continuous progress of seismic exploration technique and the application of multiple tracks multicomponent seismic survey acquisition technique, make the data volume of seismic prospecting huge, this brings certain difficulty to the Storage and Processing of geological data.In geological data storing process, geological data is record a kind of special file layout of composition (as SEG-Y, SEG-D file layout etc.) in units of seismic trace generally, its geological data message structure stored as shown in Figure 1, it is made up of file volume header and seismic trace information, and wherein seismic trace information comprises seismic trace header and seismic channel data two parts.Multiple seismic channel set can be obtained from seismic data acquisition according to Juan Touhe road header, seismic channel set is the set of the seismic trace in geological data with a certain predicable, it is the basic module in seismic data process, and such as conventional seismic channel set has common-shot-gather (CSG), common receiving point gather (CRG), common midpoint gather (CMP) and common offset road collection (COG).Usual seismic data acquisition is according to a certain seismic channel set mode record in order, therefore the seismic trace in this seismic channel set is continuous print (referred to as continous way seismic channel set) in seismic data acquisition, seismic trace now in other several seismic channel set just cannot keep continuous, and be in a discrete distribution (referred to as discontinuous seismic channel set) in seismic data acquisition.According to the different attribute of seismic trace, obtain different seismic channel set combinations from the seismic trace of the M Fig. 1, be called that collection extracts.
When earthquake data storage medium is disk, in extraction discontinuous seismic channel set process, because seismic trace position to be extracted is discontinuous, therefore after disk read a seismic channel data, magnetic head needs Mechanical Moving to arrive the position of next seismic channel data, just can restart read data operation, namely the access process of disk belongs to random access.During visit data, response time of disk by magnetic head seek time, rotate delay time and delivery time three part forms.The seek time representative value of main flow disk is 4.6ms in the market, rotates the rotating speed that disk is then depended in time delay: the rotation of common 7200RPM hard disk postpones to be 4.2ms, and that high-end 10000RPM is 3ms.Delivery time comprises the internal data delivery time and external data transmits the time, and wherein the internal data delivery time refers to after the addressing work of magnetic disk memory completes, the data time between magnetic head to hard disk cache; It is the time of computer by data bus data streams read from disk inner buffer district that external data transmits the time.The internal data transfer rate of current main flow disk is at 70 ~ 90MB/s, and external data transmission speed is at more than 100MB/s.For seismic trace, generally individually shake track data amount all not quite, be about 30KB, its delivery time is 0.3 ~ 0.4ms.Therefore, disk is in small data read-write process, and seek time and rotation time delay are much larger than the delivery time.Visible a large amount of random decimal, according to access, causes the performance of mechanical disk greatly to decline, and becomes the I/O bottleneck extracting discontinuous seismic channel set.
In actual seismic data processing, seismic data process needs process number with the mass data of TB easily, when carrying out mathematics method to geological data as process such as static correction, superposition, DMO, filtering, need in processing procedure repeatedly to extract different seismic channel set data.Therefore seismic data process has that the data volume of process is large, processing procedure is many and the feature such as process repeatedly, these features bring huge magnetic disc i/o visit capacity problem, therefore, the I/O performance of memory device becomes a key factor in mass seismic data processing time.
To sum up analyze, to reflect face key issue Seismic Exploration Data Processing system from the feature of geological data Storage and Processing: the time scale that the time that magnetic disc i/o spends accounts for total process is increasing.In Seismic Exploration Data Processing system, a kind of storage system realizing geological data specially is rapidly provided to seem particularly important.The plurality of advantages that disk array technology has due to himself is obtaining applying more and more widely, and it is a kind of vast capacity storage system be made up of polylith disk, has become during geological data stores the memory device commonly used.But the geological data storage system at present based on Disk Array Designing cannot overcome random decimal according to the inefficient problem of access I/O.
Conventional extracts the flow process of seismic channel set as shown in Figure 2 based on disk array, and seismic channel set request is become a series of read request reading seismic trace according to index translation by seismic data processing software, i.e. seek and the read function call of standard.After operating system receives these function calls, can send it to disk array controller, controller commands corresponding disk to carry out the read operation of seismic channel data more as the case may be.Transmission, after receiving seismic channel data, is performed next seismic channel data and reads process by seismic data processing software.Finally circulation is until after all seismic traces all correctly execute, seismic data processing software can obtain seismic channel set data.Flow process is extracted as can be seen from the seismic channel set of routine, because read function is generally obstruction and calls, so only after the access of current seismic track data terminates, could initiate next seismic trace director data and read, be namely each seismic channel data read in a serial fashion in seismic channel set.
According to above-mentioned flow process, analyze the performance extracting continous way seismic channel set: on disk, read the time T required for a blocks of data totalcan be described as
T total=t seek+t delay+t s(1)
In formula, t seekfor seek time, t dclayfor rotating time delay, t sfor the delivery time.
When extraction order seismic channel set, because target data is continuous print, so the read ahead technique of file system and many I/O folding can make t seek=0 and t delay=0, then formula (1) can be reduced to
T total = t s = b r D - - - ( 2 )
In formula, b rfor the data volume read and write at every turn, D is data transfer rate.
Suppose the non-redudant array structure having N block disk to form, seismic trace (establishing data volume to be about a B) distributed store is on different disks, on every block disk, seismic channel data amount is b (i) (i=1,2 ..., N), then b (1)+b (2)+...+b (N)=B, read in a seismic channel data process in order, have N block disk to walk abreast and carry out I/O operation, so disk array reads a seismic channel data time T one_seqdepend on each disk response time T totalmaximal value, be
T one _ seq = max 1 ≤ i ≤ N T total ( i ) = max 1 ≤ i ≤ N ( b ( i ) D ) - - - ( 3 )
Suppose in the ideal case, seismic channel data amount is uniformly distributed and is stored on each piece of disk, then b (i)=B/N, so reads a sequential seismic trace shortest time and be T s_min=B/ND.When containing G seismic trace in continous way seismic channel set, the shortest time T of its disk array extraction order seismic channel set g_seqfor
T G _ seq = GT s _ min = GB ND - - - ( 4 )
Formula (4) can be found out, disk array can extract continous way seismic channel set fast parallelly, and this is the major reason that disk array becomes during geological data stores the memory device commonly used.
In like manner, extract the performance of discontinuous seismic channel set: as previously mentioned according to above-mentioned conventional disk array system analysis, the process that disk is extracting discontinuous seismic channel set belongs to random decimal according to access, therefore t seekand t delaymuch larger than t s.The time that disk array reads a seismic trace depends on each disk response time T totalmaximal value, known according to formula (1), read a Distributed seismic road time T one_randfor
T one _ rand = max 1 ≤ i ≤ N T total ( i ) ≈ max 1 ≤ i ≤ N ( t seek ( i ) + t delay ( i ) ) , i = 1,2 , · · · , N - - - ( 5 )
When under disk model same case each in disk array, visual each piece of disk t average seek time scek_averagetime delay t is rotated with average delay_avcrageapproximately equal, then formula (5) can be reduced to
T one_rand=t seek_averge+t delay_averge(6)
From upper analysis, the conventional seismic channel set extraction process based on disc array system reads seismic trace in a serial fashion, therefore, when containing G seismic trace unit in discontinuous seismic channel set, then extracts time T g_randfor
T G_rand=GT one_rand=G(t seek_averge+t delay_averge) (7)
Formula (7) shows, the time extracting discontinuous seismic channel set mainly limits by the mechanical property of disk, has nothing to do with number of disks.As can be seen here, the I/O access efficiency of discontinuous seismic channel set cannot be improved by the mode of expansion disk array number of disks.To sum up, from formula (4) and formula (7), although conventional disc array system can improve the I/O access efficiency of continous way seismic channel set, the extraction speed of discontinuous seismic channel set cannot be improved.
Solid state hard disc (Solid State Disk, SSD) is the new technology developed in recent years, which overcomes the mechanical property restriction of mechanical hard disk, has good random access efficiency.But solid state hard disc memory space little (commercial solid state hard disc capacity is less than 512GBytes mostly) and expensive, the mass seismic data be not suitable for stores.Secondly, because solid state hard disc adopts Flash particle memory technology, this technology will carry out erase operation to this unit before certain unit of write, and Flash has erasing times to limit, and repeatedly after erasing, can cause storage-unit-failure.Current Large Copacity SSD uses MLC (Multi-Level Cell) technology, and the erasable number of times that each unit can be supported is at about 1000 times.In seismic data process, need the write of a large amount of data, in this case, the impact that the at every turn erasable loss of SSD brings is just particularly evident.In sum, SSD is when mass data frequently writes, and serviceable life can shorten, and likely can cause loss of data, so the drawback such as high cost, short life that SSD exists makes it be unsuitable for being used as the storage of geological data.
Summary of the invention
The object of the present invention is to provide a kind of seismic channel set paralleling abstracting method of geological data, solve the problem that existing seismic channel set extraction efficiency is low and the large cost of storage system is high.
For solving the problem, the technical solution used in the present invention is:
A seismic channel set paralleling abstracting method for geological data, comprises the following steps:
Step one, uses in homogenising memory technology write disk by seismic data acquisition by seismic channel data;
Step 2, after the seismic data processing software in client sends and reads seismic channel set request instruction, seismic channel set instruction is sent on each back end with broadcast mode;
Step 3, back end is resolved the seismic channel set instruction received, and is converted to and reads seismic trace request queue accordingly, then reads the seismic channel data in local disk according to queue, finally the seismic channel data of reading is returned to client;
Step 4, client receives the seismic channel data that back end array returns, and presses the seismic trace order sequence of collection to seismic channel data, finally returns to seismic data processing software.
Further, in described step one by seismic data acquisition by the concrete grammar in seismic channel data homogenising memory technology write disk be:
Client obtains each back end information, comprises back end total quantity N and each back end IP address;
Initialization being carried out to back end, in each back end, namely sets up corresponding filename or numbering correspondence with it, for each data track pseudorandomcode mode determines memory location, being convenient to search when reading data;
Client sends order expression to back end array will writing in files, after back end receives orders, starts preparation and accepts packet;
In units of seismic trace, read seismic data acquisition, according to seismic channel data homogenising memory technology, seismic channel data is sent to back end in the mode of network packet respectively, back end is write in corresponding file after receiving network packet;
After seismic data acquisition is sent completely, send write end of file order to back end, written document process terminates.
Further, described back end, according to resolving the seismic trace queue generated, judges whether seismic trace is stored on local disk, if exist, then reads seismic channel data, and is back to client; If do not exist, then perform next request in queue, so circulation is until queue terminates.
Further, in described step 3, back end is after being resolved the seismic channel set instruction received by MCU, corresponding seismic trace request queue is converted into according to index file, back end reads seismic channel data according to the task in request queue, then is that a networking command bag returns to client by this data encapsulation; In described step 4, client presses the seismic trace order sequence of collection to seismic channel data, and it is consistent with the sequence of request queue along row.
Further, the form of described seismic data acquisition is SEG-Y form, described SEG-Y form preserves geological data with CSG sequential system record, extracts speed for tested object, and be analyzed with the RAIDO storage system of routine with discontinuous seismic channel sets such as CRG, CMP and COG; Using extraction seismic trace number per second as the performance testing index of equipment.
The beneficial effect adopting technique scheme to produce is: solve routine based on disc array system because serial extracts seismic channel set thus the problem of the inefficiency caused, paralleling abstracting seismic channel set can be realized, namely this storage system is when there being N block disk, I/O access rate can reach approximate N lifting doubly, thus substantially increase the I/O efficiency of geological data, and extensibility is strong; The time allowing each back end execute the instruction of discontinuous seismic channel set reaches unanimity, and ensure that storage system can reach the extraction efficiency of expected in theory.
Accompanying drawing explanation
Fig. 1 is typical seismic data acquisition form schematic diagram.
Fig. 2 is the conventional schematic flow sheet extracting seismic channel set based on disc array system.
Fig. 3 is the system architecture schematic diagram of fast parallel storage system of the present invention.
Fig. 4 is the schematic flow sheet of paralleling abstracting seismic channel set of the present invention.
Fig. 5 is that seismic data acquisition of the present invention stores schematic diagram.
Fig. 6 is memory device casing structure schematic diagram of the present invention.
To be method provided by the invention and conventional memory method scheme extracting CRG seismic channel set velocity contrast Fig. 7, and solid squares body surface shows and utilizes method of the present invention, and filled circles represents conventional RAIDO method.
To be method provided by the invention and conventional memory method scheme extracting COG seismic channel set velocity contrast Fig. 8, and solid squares body surface shows and utilizes method of the present invention, and filled circles represents conventional RAIDO method.
To be method provided by the invention and conventional memory method scheme extracting CMP seismic channel set velocity contrast Fig. 9.Solid squares body surface shows and utilizes method of the present invention, and filled circles represents conventional RAIDO method.
Figure 10 adopts the seismic channel set of the inventive method to extract speed with number of disks variation relation figure.
In figure: 1, client computer, 2, gigabit ethernet switch, 3, disk Controlling vertex, 4, disk, 5, reference power supply, 6, disk Controlling vertex array, 7, ethernet interface module, 8, external interface, 9, cooling system, 10, disk.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
As shown in Figure 3, every block disk is arranged on based on composition data node on the Controlling vertex hardware of MCU, the extraction task of seismic channel set is jointly completed by N number of back end, back end array is connected to client by Ethernet, and client is responsible for sending instruction to back end and receiving the data of preserving back end array and returning.
Fig. 4 combined cover Fig. 3 shows an embodiment of the seismic channel set paralleling abstracting method of a kind of geological data of the present invention: a kind of seismic channel set paralleling abstracting method of geological data, comprises the following steps:
Step one, uses in homogenising memory technology write disk by seismic data acquisition by seismic channel data;
Step 2, after the seismic data processing software in client sends and reads seismic channel set request instruction, seismic channel set instruction is sent on each back end with broadcast mode;
Step 3, back end is resolved the seismic channel set instruction received, and is converted to and reads seismic trace request queue accordingly, then reads the seismic channel data in local disk according to queue, finally the seismic channel data of reading is returned to client;
Step 4, client receives the seismic channel data that back end array returns, and presses the seismic trace order sequence of collection to seismic channel data, finally returns to seismic data processing software.
Because back end can resolve seismic channel set instruction independently, and then corresponding seismic channel data in local disk can be read according to road collection information.Therefore, seismic channel set instruction is equivalent to by multiple back end executed in parallel.Which solves routine based on disc array system because serial extracts seismic channel set thus the problem of the inefficiency caused.
Described back end performs the flow process of seismic trace queue, it is characterized in that back end is according to resolving the seismic trace queue generated, and judges whether seismic trace is stored on local disk, if exist, then reading seismic channel data, and being back to client; If do not exist, then perform next request in queue, so circulation is until queue terminates.
According to a preferred embodiment of the seismic channel set paralleling abstracting method of a kind of geological data of the present invention, in described step one by seismic data acquisition by the concrete grammar in seismic channel data homogenising memory technology write disk be:
Client obtains each back end information, comprises back end total quantity N and each back end IP address;
Initialization being carried out to back end, in each back end, namely sets up corresponding filename or numbering correspondence with it, for each data track pseudorandomcode mode determines memory location, being convenient to search when reading data;
Client sends order expression to back end array will writing in files, after back end receives orders, starts preparation and accepts packet;
In units of seismic trace, read seismic data acquisition, according to seismic channel data homogenising memory technology, seismic channel data is sent to back end in the mode of network packet respectively, back end is write in corresponding file after receiving network packet;
After seismic data acquisition is sent completely, send write end of file order to back end, written document process terminates.
According to another preferred embodiment of the seismic channel set paralleling abstracting method of a kind of geological data of the present invention, described back end, according to resolving the seismic trace queue generated, judges whether seismic trace is stored on local disk, if exist, then read seismic channel data, and be back to client; If do not exist, then perform next request in queue, so circulation is until queue terminates.
In addition, performance can be extracted to discontinuous seismic channel set in the present invention to analyze: suppose that in discontinuous seismic channel set to be extracted, seismic trace number is G, the seismic trace number be distributed on every block disk is respectively g (i) (i=1,2, N), then g (1)+g (2)+... + g (N)=G.Known according to analyzing above, in reading discontinuous seismic channel data process, the seek time of disk and rotation time delay, much larger than the delivery time, under not considering cache hit situation, read single Distributed seismic road T averaging time one_avergecotype (6).Therefore each back end executes the time T of queue node_randthe seismic trace number read to this locality is directly proportional, namely
T node(i)=T one_avergeg(i) (8)
As can be seen from formula (8), when seismic trace number g (i) on each piece of disk does not wait, the time that each back end executes seismic channel set instruction is different.According to Bucket Principle, back end array completes time of extracting discontinuous seismic channel set and determines by executing seismic channel set instruction time the longest that back end, therefore extracts discontinuous seismic channel set required time T new_randfor
T new _ rand = max 1 ≤ i ≤ N T node ( i ) - - - ( 9 )
According to formula (8) and formula (9) known, as g (i)=G/N, extract the shortest time of discontinuous seismic channel set and be
T rand _ min = GT one _ averge N - - - ( 10 )
Contrast (7) are known, and the extraction discontinuous seismic channel set time of the present invention decreases N doubly than conventional disc array system in theory.
Known according to formula (9), when seismic trace number g (i) that each piece of disk stores is uneven, to the read work amount between different disk be caused to differ greatly, so extract seismic channel set required time and store the maximum disk of data by being limited to.Although the position of discontinuous seismic channel set in seismic data acquisition is dispersion, they still may have certain relevance in space distribution, namely not necessarily can be randomly dispersed in seismic data acquisition.If store seismic channel data according to plain sequence mode, probably appear at the uneven situation of Distributed seismic road number g (i) storage and distribution in each disk, this will cause efficiency of the present invention can not reach theoretic expection efficiency.
Based on this, another object of the present invention is to provide a kind of seismic channel data homogenising memory technology, if each R/W speed of hard disc is identical and storage data volume is identical, when discontinuous seismic channel set be uniformly distributed be stored on each piece of disk time, seismic trace number g (i) between different disk is almost equal, known according to formula (8), the time that each back end executes the instruction of discontinuous seismic channel set reaches unanimity, and ensure that this storage system can reach the extraction efficiency of expected in theory.
For the invention allows for seismic channel data homogenising memory technology, this technology as shown in Figure 5, after seismic data acquisition is split in units of seismic trace, produce pseudo-random integer sequence again and the seismic trace in seismic data acquisition is carried out random coded, each seismic trace unit is sent to different back end according to coding.So just can make spatially have certain relevance seismic trace can evenly be stored on different nodes as much as possible.In seismic channel set extraction process, utilize identical seed to produce identical pseudo-random sequence and carry out the memory location that decoding can obtain sympathetic earthquakes track data.Seismic channel data homogenising memory technology improves the uniformity coefficient of discontinuous seismic channel set in each piece of disk, effectively ensure that discontinuous seismic channel set extraction efficiency can reach theoretical value.
Although the seismic channel data homogenising memory technology that the present invention adopts makes continous way seismic channel set store the unordered of change to a certain extent, but the extensibility of the number of disks of this system will far above disk array, by expanding the mode of number of disks, can improve the I/O access efficiency of continous way seismic channel set, therefore the I/O access efficiency of the present invention when extracting continous way seismic channel set is also not less than conventional disc array system.
According to another preferred embodiment of the seismic channel set paralleling abstracting method of a kind of geological data of the present invention, in described step 3, back end is after being resolved the seismic channel set instruction received by MCU, corresponding seismic trace request queue is converted into according to index file, back end reads seismic channel data according to the task in request queue, then is that a networking command bag returns to client by this data encapsulation; In described step 4, client presses the seismic trace order sequence of collection to seismic channel data, and it is consistent with the sequence of request queue along row.
According to another preferred embodiment of the seismic channel set paralleling abstracting method of a kind of geological data of the present invention, the form of described seismic data acquisition is SEG-Y form, described SEG-Y form preserves geological data with CSG sequential system record, extract speed for tested object with CRG, CMP and COG discontinuous seismic channel set, and be analyzed with the RAIDO storage system of routine; Using extraction seismic trace number per second as the performance testing index of equipment, its relational expression is
TPS = G T ( trace / s ) - - - ( 11 )
In formula, G is the number of seismic trace in seismic channel set, and T is the T.T. extracting seismic channel set use.
For earthquake memory device cabinet, its concrete layout as shown in Figure 6, comprising: reference power supply 5, disk Controlling vertex array 6, ethernet interface module 7, external interface 8, cooling system 9, redundant array of inexpensive disks 10.The fan that reference power supply 5 is supplied in back end (5V), ethernet interface module (12V), hard disk (5V, 12V) and cooling system is powered, and back end 6 adopts the development board based on Linux.What ethernet interface module 7 adopted is 8 port gigabit ethernet switches.In ethernet module, 6 ports are connected with 6 back end, other 2 ethernet ports are as external interface 8, external interface 8 is connected on gigabit ethernet switch by netting twine, what disk was selected is Seagate ST1000DM003,1TB capacity, 7200 turns of hard disks, and disk is connected to disk Controlling vertex by SATA interface.Client computer is configured with 2 network interface cards and is connected on gigabit ethernet switch respectively by network, and one of them network interface card is as DHCP service, and be namely back end respectively IP address by DHCP agreement, another network interface card is used for the transmission channel of geological data.In client computer, a seismic data processing software is installed under Linux environment.
Fig. 7,8 and 9 is respectively the extraction velocity contrast design sketch that 3 kinds of discontinuous seismic channel sets utilize method provided by the invention and conventional storage mode, wherein Fig. 7 is that totally 8475 roads extract velocity contrast figure in CRG, Fig. 8 is that totally 108917 roads extract velocity contrast figure in COG, and Fig. 9 is that totally 15156 roads extract velocity contrast figure in CMP.Solid squares body surface shows and utilizes method of the present invention, and filled circles represents conventional RAIDO method.
Test result in comparison diagram can be found out, seismic channel set provided by the invention is utilized to extract speed method, it extracts speed TPS and number of disks is good linear, and the extraction speed TPS of the RAIDO system of routine increases progressively without marked change with number of disks, and speed TPS is almost constant in its extraction.Simultaneously, when disk number is N=6, the present invention extracts speed TPS far away higher than the RAIDO systems approach of routine, this has absolutely proved that equipment provided by the invention and method effectively improve geological data and extract speed, solves conventional based on the inefficient problem of disc array system extraction seismic channel set.
Figure 10 is that the extraction speed TPS that utilizes seismic channel set abstracting method provided by the invention to test to obtain is with number of disks variation relation schematic diagram, as can be seen from the figure, when number of disks is increased to more than 20, extraction speed TPS still remains good linear relationship, shows that the inventive method extendability is strong.
Although with reference to multiple explanatory embodiment of the present invention, invention has been described here, but, should be appreciated that, those skilled in the art can design a lot of other amendment and embodiment, these amendments and embodiment will drop within spirit disclosed in the present application and spirit.More particularly, in the scope of, accompanying drawing open in the application and claim, multiple modification and improvement can be carried out to the building block of subject combination layout and/or layout.Except the modification of carrying out building block and/or layout is with except improvement, to those skilled in the art, other purposes also will be obvious.

Claims (5)

1. a seismic channel set paralleling abstracting method for geological data, is characterized in that comprising the following steps:
Step one, uses in homogenising memory technology write disk by seismic data acquisition by seismic channel data;
Step 2, after the seismic data processing software in client sends and reads seismic channel set request instruction, seismic channel set instruction is sent on each back end with broadcast mode;
Step 3, back end is resolved the seismic channel set instruction received, and is converted to and reads seismic trace request queue accordingly, then reads the seismic channel data in local disk according to queue, finally the seismic channel data of reading is returned to client;
Step 4, client receives the seismic channel data that back end array returns, and sorts to the seismic trace order of seismic channel data by road collection, finally returns to seismic data processing software.
2. the seismic channel set paralleling abstracting method of a kind of geological data according to claim 1, it is characterized in that: in described step one by seismic data acquisition by the concrete grammar in seismic channel data homogenising memory technology write disk be: client obtains each back end information, comprises back end total quantity N and each back end IP address; Initialization being carried out to back end, in each back end, namely sets up corresponding filename or numbering correspondence with it, for each data track pseudorandomcode mode determines memory location, being convenient to search when reading data; Client sends order expression to back end array will writing in files, after back end receives orders, starts preparation and accepts packet;
In units of seismic trace, read seismic data acquisition, according to seismic channel data homogenising memory technology, seismic channel data is sent to back end in the mode of network packet respectively, back end is write in corresponding file after receiving network packet;
After seismic data acquisition is sent completely, send write end of file order to back end, written document process terminates.
3. the seismic channel set paralleling abstracting method of a kind of geological data according to claim 1, it is characterized in that: described back end, according to resolving the seismic trace queue generated, judges whether seismic trace is stored on local disk, if exist, then read seismic channel data, and be back to client; If do not exist, then perform next request in queue, so circulation is until queue terminates.
4. the seismic channel set paralleling abstracting method of a kind of geological data according to claim 1, it is characterized in that: in described step 3, back end is after being resolved the seismic channel set instruction received by MCU, corresponding seismic trace request queue is converted into according to index file, back end reads seismic channel data according to the task in request queue, then is that a networking command bag returns to client by this data encapsulation; In described step 4, client presses the seismic trace order sequence of collection to seismic channel data, and it is consistent with the sequence of request queue along row.
5. the seismic channel set paralleling abstracting method of a kind of geological data according to claim 1-4 any one, it is characterized in that: the form of described seismic data acquisition is SEG-Y form, described SEG-Y form preserves geological data with CSG sequential system record, extract speed for tested object with CRG, CMP and COG discontinuous seismic channel set, and be analyzed with the RAIDO storage system of routine; Using extraction seismic trace number per second as the performance testing index of equipment.
CN201510073212.7A 2015-02-11 2015-02-11 A kind of seismic channel set paralleling abstracting method of geological data Active CN104570063B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510073212.7A CN104570063B (en) 2015-02-11 2015-02-11 A kind of seismic channel set paralleling abstracting method of geological data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510073212.7A CN104570063B (en) 2015-02-11 2015-02-11 A kind of seismic channel set paralleling abstracting method of geological data

Publications (2)

Publication Number Publication Date
CN104570063A true CN104570063A (en) 2015-04-29
CN104570063B CN104570063B (en) 2017-10-10

Family

ID=53086577

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510073212.7A Active CN104570063B (en) 2015-02-11 2015-02-11 A kind of seismic channel set paralleling abstracting method of geological data

Country Status (1)

Country Link
CN (1) CN104570063B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106291688A (en) * 2016-07-25 2017-01-04 中国石油天然气股份有限公司 A kind of processing method and processing device of post-stack seismic data
CN106790389A (en) * 2016-11-25 2017-05-31 中国石油天然气集团公司 The acquisition methods of seismic channel data, host node server and working node server
CN107094085A (en) * 2016-02-17 2017-08-25 中国移动通信集团江苏有限公司 A kind of signaling methods and device
CN109343117A (en) * 2018-11-10 2019-02-15 北京科胜伟达石油科技股份有限公司 Double buffer dual-thread seismic data display methods
CN109655911A (en) * 2017-10-11 2019-04-19 中国石油化工股份有限公司 Seismic data visualization system and method based on WebService
CN110187385A (en) * 2019-06-27 2019-08-30 北京中油瑞飞信息技术有限责任公司 Seismic data acquisition method, seismic data processing technique and device
CN111399776A (en) * 2020-03-13 2020-07-10 西安石油大学 Geological exploration management system based on data acquisition
CN117376289A (en) * 2023-10-11 2024-01-09 哈尔滨工业大学 Network disk data scheduling method for road monitoring data use application
CN117519610A (en) * 2024-01-05 2024-02-06 中国石油集团东方地球物理勘探有限责任公司 Method, device and medium for creating CMP data body by post-stack seismic data

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002103397A1 (en) * 2001-06-16 2002-12-27 Westerngeco Seismic Holdings Limited A method of processing data
CN101839997A (en) * 2009-03-18 2010-09-22 中国石油天然气集团公司 Seismic data parallel storage system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002103397A1 (en) * 2001-06-16 2002-12-27 Westerngeco Seismic Holdings Limited A method of processing data
CN101839997A (en) * 2009-03-18 2010-09-22 中国石油天然气集团公司 Seismic data parallel storage system

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
刘永江 等: "基于道集流的地震数据并行输入输出方法", 《东北石油大学学报》 *
文必龙 等: "虚拟地震数据文件并行访问策略", 《计算机系统应用》 *
王意洁 等: "云计算环境下的分布存储关键技术", 《软件学报》 *
赵满: "地震数据并行访问策略的研究", 《中国优秀硕士学位论文全文数据库 基础科技辑》 *
赵长海 等: "面向地震数据处理的并行与分布式编程框架", 《石油地球物理勘探》 *

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107094085B (en) * 2016-02-17 2020-05-12 中国移动通信集团江苏有限公司 Signaling transmission method and device
CN107094085A (en) * 2016-02-17 2017-08-25 中国移动通信集团江苏有限公司 A kind of signaling methods and device
CN106291688B (en) * 2016-07-25 2019-02-15 中国石油天然气股份有限公司 A kind of processing method and processing device of post-stack seismic data
CN106291688A (en) * 2016-07-25 2017-01-04 中国石油天然气股份有限公司 A kind of processing method and processing device of post-stack seismic data
CN106790389A (en) * 2016-11-25 2017-05-31 中国石油天然气集团公司 The acquisition methods of seismic channel data, host node server and working node server
CN109655911A (en) * 2017-10-11 2019-04-19 中国石油化工股份有限公司 Seismic data visualization system and method based on WebService
CN109343117A (en) * 2018-11-10 2019-02-15 北京科胜伟达石油科技股份有限公司 Double buffer dual-thread seismic data display methods
CN110187385A (en) * 2019-06-27 2019-08-30 北京中油瑞飞信息技术有限责任公司 Seismic data acquisition method, seismic data processing technique and device
CN110187385B (en) * 2019-06-27 2022-03-01 北京中油瑞飞信息技术有限责任公司 Seismic data acquisition method, seismic data processing method and device
CN111399776A (en) * 2020-03-13 2020-07-10 西安石油大学 Geological exploration management system based on data acquisition
CN117376289A (en) * 2023-10-11 2024-01-09 哈尔滨工业大学 Network disk data scheduling method for road monitoring data use application
CN117376289B (en) * 2023-10-11 2024-04-12 哈尔滨工业大学 Network disk data scheduling method for road monitoring data use application
CN117519610A (en) * 2024-01-05 2024-02-06 中国石油集团东方地球物理勘探有限责任公司 Method, device and medium for creating CMP data body by post-stack seismic data
CN117519610B (en) * 2024-01-05 2024-03-22 中国石油集团东方地球物理勘探有限责任公司 Method, device and medium for creating CMP data body by post-stack seismic data

Also Published As

Publication number Publication date
CN104570063B (en) 2017-10-10

Similar Documents

Publication Publication Date Title
CN104570063A (en) Parallel extraction method of seismic channel set of seismic data
US10996883B2 (en) Storage system including host device and storage device configured to perform selective purge operation
US9996435B2 (en) Reliability scheme using hybrid SSD/HDD replication with log structured management
KR101852668B1 (en) Address mapping
CN106548789B (en) Method and apparatus for operating stacked tile type magnetic recording equipment
US10725930B2 (en) Logical to physical memory address mapping tree
CN101676882B (en) Built-in mapping message of memory device
CN106708424A (en) Apparatus and method for performing selective underlying exposure mapping on user data
CN106708751A (en) Storage device including multi-partitions for multimode operations, and operation method thereof
CN107346210B (en) Hard disk data erasing method, server and system
JP5891027B2 (en) Method and apparatus for sanitizing a storage device
CN101840308A (en) Hierarchical memory system and logical volume management method thereof
TW201526003A (en) Data storage device and flash memory control method
CN104461397A (en) Solid-state drive and read-write method thereof
TW201005536A (en) Solid state storage system with data attribute wear leveling and method of controlling the solid state storage system
CN109164975A (en) A kind of method and solid state hard disk writing data into solid state hard disk
CN103500146A (en) Method and system for migrating stored data of virtual machine disks
US20240143219A1 (en) Software-hardware combination method for internal mapping address query of zoned namespace
CN104270409A (en) Method for efficiently utilizing centralized storage in cloud platform
CN104898989B (en) A kind of Mass Data Storage Facility, method and device
CN107577549A (en) It is a kind of to store the method for testing for deleting function again
KR101480424B1 (en) Apparatus and method for optimization for improved performance and enhanced lifetime of hybrid flash memory devices
US10929066B1 (en) User stream aware file systems with user stream detection
CN106547472A (en) Storage array management method and device
CN103617010A (en) Data storage method and storage device

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right

Effective date of registration: 20170823

Address after: Baohe District of Hefei City, Anhui province 230000 Huizhou Avenue International room 2419

Applicant after: Hefei keen Storage Technology Co., Ltd.

Address before: 616, building A3, building 800, innovation industrial park, Wangjiang West Road, Hefei hi tech Zone, Anhui 230000, China

Applicant before: ANHUI JITUO ELECTRONIC TECHNOLOGY CO., LTD.

TA01 Transfer of patent application right
GR01 Patent grant
GR01 Patent grant