CN107479833A - Key value storage-oriented remote nonvolatile memory access and management method - Google Patents
Key value storage-oriented remote nonvolatile memory access and management method Download PDFInfo
- Publication number
- CN107479833A CN107479833A CN201710716667.5A CN201710716667A CN107479833A CN 107479833 A CN107479833 A CN 107479833A CN 201710716667 A CN201710716667 A CN 201710716667A CN 107479833 A CN107479833 A CN 107479833A
- Authority
- CN
- China
- Prior art keywords
- memory
- memory block
- data
- area
- client
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0679—Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
Abstract
The invention relates to a remote nonvolatile memory access and management method facing key value storage, which is characterized in that a nonvolatile memory cache region is set for a key value storage system and is registered on a network card capable of supporting remote direct memory access technology; this non-volatile memory cache area is divided into two parts: the memory blocks in the sparse area are mainly used for receiving data of a remote client, different memory blocks are allocated for remote data writing each time, the memory block in each sparse area only belongs to one key value pair, the memory block in the compact area is a main cache area of the key value pair, each memory block comprises a plurality of key value pairs, and a read request of the client is served; compressing key value pairs in a plurality of allocated memory blocks in the sparse area into memory blocks in one or more compact areas by a compressor at regular intervals by the server; when the number of available memory blocks of the compact area is insufficient, data is replaced to the data storage area. The invention can relieve the fixed point write abrasion problem of the remote nonvolatile memory and improve the remote read-write performance.
Description
Technical field
It is particularly a kind of towards the remote of key assignments storage the present invention relates to a kind of long-range nonvolatile memory access and management method
Journey nonvolatile memory accesses and management method.
Background technology
, it is necessary to face the storage demand of mass data under big data environment.On the one hand, the autgmentability of application system is carried
Go out higher requirement, and the NoSQL systems based on key assignments storage architecture, the storage of data is using independent key-value pair to be substantially single
Member, the relevance between data is reduced, there is good autgmentability；On the other hand, the meter based on convential memory and external memory framework
Calculation machine storage system is faced with the challenge of mass data processing performance；Novel memory devices part, as PCM (phase-change random access memory),
3D Xpoint, possess the characteristics such as the access speed of the non-volatile of external memory, high integration and internal memory level, it is contemplated that existing
The deficiency of DRAM (dynamic random access memory) volatile characteristic, high energy consumption and finite capacity, formed based on these devices
Nonvolatile memory (NVM, Non-Volatile Memory) turn into Computer Design and researcher and improve performance of storage system
Hot topic.Although NVM integrated level is higher than DRAM, the NVM capacity of unit or limited and current NVM skills
The maturity of art is not high, reports and shows according to Intel：In the computer system in future, 4 times of convential memory capacity will be disposed
It is commercialized NVM products (such as 3D Xpoint)；Really, the NVM resources of unit are held far from the storage for meeting big data application system
Amount demand, thus by network it is extending transversely become it is necessary.Legacy network based on ICP/IP protocol can turn into long-distance inner
The performance bottleneck of extension, main cause are the data processing methods of ICP/IP protocol, buffer area of the application program in user's space
The data being ready for, then called the kernel spacing of data copy by system, needed in kernel spacing to transmission
Data add the header information needed for each layer protocol, and the machine of distal end is finally sent to by network interface card, and the machine of distal end passes through net
Clamping receives data, according to reverse logical analysis packet is sent to, finally by the use of data copy to remote application
The buffer area in family space；It is seen that carried out data transmission using conventional network techniques, it is necessary to undergo multiple data copy,
The copy of data not only consumes CPU resource, also reduces the utilization rate of network bandwidth.Compared with TCP/IP, RDMA
(Remote Direct Memory Access, long-range directly to access memory techniques) allows remotely directly to access long-range user
Memory headroom, without distal end CPU participation, during data transfer, the CPU of distal end can run other journeys parallel
Sequence.Current remote access internal memory be based on DRAM, due to NVM internal storage access characteristic, remote access NVM be it is rational,
And NVM will play the part of important role in the RDMA technologies in future.
But, while NVM possesses the advantages characteristic of internal memory and external memory, NVM also there is read and write it is unbalanced, write abrasion
The problems such as, because RDMA technologies can allow the nonvolatile memory of distal end directly to be accessed by client, and RDMA is accessed
The mapping of virtual address and physical address be fixed, this has speeded up NVM and has write loss, so, how to design NVM access
Mode and administrative mechanism become new challenge.
The content of the invention
In order to solve above-mentioned technical problem, it is an object of the invention to provide a kind of towards the long-range non-easy of key assignments storage
Lose internal storage access and management method.
This method is key assignments storage one nonvolatile memory buffer area of setting, and is registered to and can support remote direct memory
On the network interface card of access technique；It is two parts to divide this nonvolatile memory buffer area：Rarefaction and compact area, in rarefaction
Memory block is mainly used in receiving the data of Terminal Server Client, and is write per the different memory block of sub-distribution for teledata,
The memory block of each rarefaction is pertaining only to a key-value pair, and it is to improve the performance of remote write and extend non-volatile that it, which is acted on,
The internal memory life-span.
And the memory block in compact area is the main buffer area of key-value pair, each memory block includes multiple key-value pairs, service visitor
The read request at family end；Service end is every at regular intervals, by compression mechanism by the multiple allocated memory blocks in rarefaction
Key-value pair is compressed in the memory block in one or more compact areas；When the lazy weight of the free memory block in compact area, can incite somebody to action
Data are replaced to data storage area.It is the efficiency for improving caching that it, which is acted on,.
The technical solution adopted for the present invention to solve the technical problems is：
A kind of long-range nonvolatile memory based on RDMA accesses and management method, including：
The first step, service end division in save as three classes：Data storage area, nonvolatile memory buffer area and command buffer；Its
In, the internal memory of data storage area and nonvolatile memory buffer area is NVM, and the internal memory of command buffer is DRAM；
Second step, the internal memory of nonvolatile memory buffer area and command buffer is registered to by service end supports remote direct memory to visit
Ask on the network interface card of technology, and the registration for the parameter selectivity that data storage area is then set according to user, and nonvolatile memory is delayed
Deposit zoning and be divided into compact area and rarefaction, the size of the memory block of rarefaction is uniformly set as the multiple of page byte, and the multiple is
1-n times, n is positive integer；
3rd step, service end receives the connection request of client, and distributes a rarefaction memory block and one for each client
The memory block of individual command buffer；
4th step, client (Remote Direct Memory Access, are remotely directly visited when transmitting data by RDMA
Ask memory techniques) the sparse memory block of write-in, while send control information in order buffer memory block, service end returns to next time
The sparse block address memory write is as write acknowledgement (ACK) message；If one time data transfer needs multiple sparse memory blocks, root
According to control information, multiple sparse memory blocks are cascaded and ensure the integrality of data；
5th step, when the quantity of the free memory block of sparse region drops to certain threshold values（The threshold value refers to：Free memory
The quantity of block and the ratio of total internal memory number of blocks, scope 0.1~0.6）, service end will will select by Memory Compress Mechanism
Data compression in multiple sparse memory blocks reclaims the sparse internal memory for including junk data into the memory block in compact area
Block；
6th step, when the quantity of the free memory block in compact area drops to certain threshold values（The threshold value refers to：Free memory block
Quantity and the ratio of total internal memory number of blocks, scope 0.1~0.6）, service end will will be selected multiple by Data Migration mechanism
Data copy in the memory block in compact area is to data storage area.
It can reach following beneficial effect using the present invention：Firstly, since limited, division clothes are cached on piece on RDMA network interface cards
The nonvolatile memory at business end be data storage area and data buffer area, log-on data buffer area and the storage of selective log-on data
Area, reduce the performance for being not hit by ensure that data manipulation while expense cached on the piece of network interface card；Secondly, by data buffer storage
Rarefaction and compact area are divided into, for client, when sending data each time, the memory block of rarefaction is different
, it means that it is not fixed point that the data of client, which are write, can so extend nonvolatile memory writes the life-span；Furthermore
Client distributes a sparse memory block as the confirmation message write complete after writing every time, ensure that the reliability of data transfer
While improve the performance that continuous data is write；Finally, the valid data of sparse memory block will be taken by Memory Compress Mechanism
It is compressed in compact area's memory block, improves the service efficiency of internal memory.
Brief description of the drawings
The invention will be further described with reference to the accompanying drawings and examples.
Fig. 1 is the access of long-range nonvolatile memory and management flow chart towards key assignments storage；
Fig. 2 is the sparse and compacting memory block organization chart in data buffer area.
Embodiment
Reference picture 1, the invention provides a kind of long-range nonvolatile memory access towards key assignments storage and management method, bag
Include:
The first step, for service end to RDMA network interface card registering communication region of memorys, communication region of memory includes nonvolatile memory and volatile
(tradition) internal memory, nonvolatile memory are used for the caching of key-value pair, and volatile memory receives the command information of user data transmission；
Second step, service end division nonvolatile memory is sparse and compact two regions, and acquiescently, the two regions are united by size
One multiple memory blocks composition, the size of the memory block of rarefaction is in 4096 bytes between 16384 bytes；
3rd step, client are initiated to establish connection request to service end, and service end is that each client distributes a reception order
Buffer memory block and a sparse region memory block, and these information are returned into client；
4th step, client are first write key-value pair in this sparse memory block by the long-range write operations of RDMA, and will control
Message processed is sent into command buffer；
5th step, service end continue to distribute a sparse region by the control message in resolve command buffering area for client
In memory block as the data that send of storage next time；If this is last time data block, then in upper once sparse
The identification information of itself is write in memory block and represents this data transfer ends, otherwise, in upper sparse memory block end once
It is written as the mark of the memory block for the sparse region that data transfer next time is distributed；
6th step, if more than one internal memory bound block size of the residue length of the key-value pair data of client, continue weight
Multiple step 4；
7th step, at interval of a period of time, the quantity of the free memory block of service end statistics sparse region, if free memory block
Data are reduced to certain threshold values, and service end selects a number of memory block, then will be therein by data compression scheme
Data copy is into the memory block in one or more compact areas；
8th step, at interval of a period of time, the quantity of the free memory block of service end statistics compact area, if free memory block
Data are reduced to certain threshold values, and service end selects a number of memory block, then will be therein by Data Migration mechanism
Data copy increases available internal memory number of blocks to data storage areas；
Preferred embodiment is further used as, described step two, it is specially：
The internal memory of the buffer area of registration is that all clients are shared, the memory block in compact area and the memory block of rarefaction of acquiescence
Size it is the same, for the memory block of rarefaction, its main effect is for caching more key-value pairs, and client is only
There is the authority of reading, manage for convenience, the size of all memory blocks in this region is unified, but to allowing single
Memory block cache more data, the size maximum of the memory block in rarefaction can be the several of compact area's internal memory block size
Times；Because each client can bind the memory block in a compact area, if the internal memory block size setting in this region is too big,
In the case where client increases, the resource consumption of internal memory is more, so, it is set as the memory block in compact area according to size
Multiple grades, such as 4096 bytes, 8192 bytes, 16384 bytes, reference picture 2, connection or each key-value pair are established in client
When transferring, a minimum memory block, that is, 4096 bytes are only bound；The distribution of other memory blocks, then according to client
The demand dynamic at end adjusts.
Preferred embodiment is further used as, described step three, it is specially：
Due to the discontinuous nature of the data sending request of client, the binding memory block fixed for a long time if each connection, then
Other can be caused to need the client of memory block to lack resource, so setting is each connected in the regular hour, if do not had
The transmission of data, then can reclaim the memory block in the compact area of binding, and recovery has two ways：Client proactive notification abandons this
The authority and service end of memory block force recovery memory block and notify client transmission next time data to need to apply for memory block again.
Claims (5)
1. a kind of long-range nonvolatile memory towards key assignments storage accesses and management method, non-volatile for key assignments storage setting one
Internal memory cache region, and be registered on the network interface card that can support remote direct memory access technique；It is characterized in that, described non-volatile
Internal memory cache region is divided into two parts：Rarefaction and compact area, the memory block in rarefaction are used to receive Terminal Server Client
Data, and write per the different memory block of sub-distribution for teledata, the memory block of each rarefaction is pertaining only to a key
Value pair, to improve the performance of remote write and extend the nonvolatile memory life-span；
The memory block in compact area is the main buffer area of key-value pair, and each memory block includes multiple key-value pairs, service client
Read request；Service end is per at regular intervals, by compression mechanism by the key-value pair in the multiple allocated memory blocks in rarefaction
It is compressed in the memory block in one or more compact areas；Improve the efficiency of caching；When compact area free memory block quantity not
When sufficient, data are replaced to data storage area；
Specific steps include：
The first step, service end division in save as three classes：Data storage area, nonvolatile memory buffer area and command buffer；Its
In, the internal memory of command buffer is DRAM；
Second step, the internal memory of nonvolatile memory buffer area and command buffer is registered to by service end supports remote direct memory to visit
Ask on the network interface card of technology, and the registration for the parameter selectivity that data storage area is set according to user；By nonvolatile memory buffer area
Compact area and rarefaction are divided into, the size of the memory block of rarefaction is uniformly set as the multiple of page byte, and the multiple is 1-n
Times, n is positive integer；
3rd step, service end receives the connection request of client, and distributes a rarefaction memory block and one for each client
The memory block of individual command buffer；
4th step, client enter sparse memory block, while send control information to order when transmitting data by RDMA Write
In buffer memory block, service end returns to the sparse block address memory write next time as write acknowledgement message；An if data transfer
Multiple sparse memory blocks are needed, then according to control information, multiple sparse memory blocks is cascaded and ensure the integrality of data；
5th step, when the quantity of the free memory block of sparse region drops to certain threshold values, service end will pass through internal memory pressure
Contracting mechanism, the data compression in multiple sparse memory blocks will be selected into the memory block in compact area, and reclaim and include rubbish number
According to sparse memory block；
6th step, when the quantity of the free memory block in compact area drops to certain threshold values, service end will pass through Data Migration
Mechanism is by the data copy in the memory block for selecting multiple compact areas to data storage area.
2. a kind of long-range nonvolatile memory towards key assignments storage according to claim 1 accesses and management method, it is special
Sign is that its effect of the memory block of the rarefaction is for caching more key-value pairs, and client only has the authority read, in order to
Convenient management, it is unified in the size of all memory blocks of rarefaction, in order to allow single memory block to cache more data,
The size maximum of memory block in rarefaction can be several times of compact area's internal memory block size；Because each client can be bound
The memory block in one compact area, if the memory block setting in compact area is too big, in the case where client increases, the resource of internal memory
Consume more, the memory block in compact area is set as several grades according to size.
3. a kind of long-range nonvolatile memory towards key assignments storage according to claim 1 accesses and management method, it is special
Sign is, the 4th step,
If without the transmission of data, the memory block in the compact area of binding is reclaimed, reclaiming has two ways：Client proactive notification
Abandon the authority of this memory block and service end forces recovery memory block and notifies client to send data needs Shen again next time
Please memory block.
4. a kind of long-range nonvolatile memory towards key assignments storage according to claim 1 accesses and management method, it is special
Sign is, the 4th step,
Client is it needs to be determined that the validity of the binding relationship of memory block before data is transmitted, if the relation of binding is solved
Remove, then the memory block of the sparse memory field of first to file, be over if service end judges that a key-value pair is sent, can be this
The transmission that a rarefaction memory block is used for key-value pair next time is bound in connection again.
5. a kind of long-range nonvolatile memory towards key assignments storage according to claim 1 accesses and management method, it is special
Sign is that the threshold value in the 5th step, the 6th step refers to：The quantity of free memory block and the ratio of total internal memory number of blocks,
Scope 0.1~0.6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710716667.5A CN107479833B (en) | 2017-08-21 | 2017-08-21 | Key value storage-oriented remote nonvolatile memory access and management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710716667.5A CN107479833B (en) | 2017-08-21 | 2017-08-21 | Key value storage-oriented remote nonvolatile memory access and management method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107479833A true CN107479833A (en) | 2017-12-15 |
| CN107479833B CN107479833B (en) | 2020-04-17 |
Family
ID=60601682
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710716667.5A Active CN107479833B (en) | 2017-08-21 | 2017-08-21 | Key value storage-oriented remote nonvolatile memory access and management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107479833B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109976947A (en) * | 2019-03-11 | 2019-07-05 | 北京大学 | A kind of method and system of the power loss recovery towards mixing memory |
| CN110535811A (en) * | 2018-05-25 | 2019-12-03 | 中兴通讯股份有限公司 | Remote memory management method and system, server-side, client, storage medium |
| CN111858418A (en) * | 2019-04-30 | 2020-10-30 | 华为技术有限公司 | Memory communication method and device based on remote direct memory access RDMA |
| CN111897784A (en) * | 2020-07-13 | 2020-11-06 | 安徽大学 | Key value storage-oriented near data computing cluster system |
| WO2022095685A1 (en) * | 2020-11-04 | 2022-05-12 | 中兴通讯股份有限公司 | Persistent memory key value system and operation method therefor |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060047771A1 (en) * | 2004-08-30 | 2006-03-02 | International Business Machines Corporation | RDMA server (OSI) global TCE tables |
| CN102843435A (en) * | 2012-09-10 | 2012-12-26 | 浪潮(北京)电子信息产业有限公司 | Access and response method and access and response system of storing medium in cluster system |
| US20140325012A1 (en) * | 2012-11-21 | 2014-10-30 | International Business Machines Corporation | Rdma-optimized high-performance distributed cache |
| CN106202138A (en) * | 2015-06-01 | 2016-12-07 | 三星电子株式会社 | Storage device and method for autonomous space compression |
| CN106469198A (en) * | 2016-08-31 | 2017-03-01 | 华为技术有限公司 | Key assignments storage method, apparatus and system |
-
2017
- 2017-08-21 CN CN201710716667.5A patent/CN107479833B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060047771A1 (en) * | 2004-08-30 | 2006-03-02 | International Business Machines Corporation | RDMA server (OSI) global TCE tables |
| CN102843435A (en) * | 2012-09-10 | 2012-12-26 | 浪潮(北京)电子信息产业有限公司 | Access and response method and access and response system of storing medium in cluster system |
| US20140325012A1 (en) * | 2012-11-21 | 2014-10-30 | International Business Machines Corporation | Rdma-optimized high-performance distributed cache |
| CN106202138A (en) * | 2015-06-01 | 2016-12-07 | 三星电子株式会社 | Storage device and method for autonomous space compression |
| CN106469198A (en) * | 2016-08-31 | 2017-03-01 | 华为技术有限公司 | Key assignments storage method, apparatus and system |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110535811A (en) * | 2018-05-25 | 2019-12-03 | 中兴通讯股份有限公司 | Remote memory management method and system, server-side, client, storage medium |
| CN110535811B (en) * | 2018-05-25 | 2022-03-04 | 中兴通讯股份有限公司 | Remote memory management method and system, server, client and storage medium |
| CN109976947A (en) * | 2019-03-11 | 2019-07-05 | 北京大学 | A kind of method and system of the power loss recovery towards mixing memory |
| CN109976947B (en) * | 2019-03-11 | 2020-11-27 | 北京大学 | Power failure recovery method and system for hybrid memory |
| CN111858418A (en) * | 2019-04-30 | 2020-10-30 | 华为技术有限公司 | Memory communication method and device based on remote direct memory access RDMA |
| CN111858418B (en) * | 2019-04-30 | 2023-04-07 | 华为技术有限公司 | Memory communication method and device based on remote direct memory access RDMA |
| CN111897784A (en) * | 2020-07-13 | 2020-11-06 | 安徽大学 | Key value storage-oriented near data computing cluster system |
| CN111897784B (en) * | 2020-07-13 | 2022-12-06 | 安徽大学 | Key value storage-oriented near data computing cluster system |
| WO2022095685A1 (en) * | 2020-11-04 | 2022-05-12 | 中兴通讯股份有限公司 | Persistent memory key value system and operation method therefor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107479833B (en) | 2020-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107479833A (en) | Key value storage-oriented remote nonvolatile memory access and management method | |
| CN103019970B (en) | Memory storage and control method thereof | |
| CN103946828B (en) | Data processing system and method | |
| CN102045772B (en) | Data transmission method and device | |
| CN101267361B (en) | A high-speed network data packet capturing method based on zero duplication technology | |
| CN101841545B (en) | TCP stream restructuring and/or packetizing method and device | |
| CN101626384B (en) | Method and system for transmitting data packet based on TPC protocol | |
| CN101751233B (en) | Method and system for expanding capacity of memory device | |
| CN102446139B (en) | Method and device for data storage | |
| US20080320155A1 (en) | Aggregation and re-ordering of input/output requests for better performance in remote file systems | |
| CN100535886C (en) | Data-transmission system between nodes, and device and method | |
| CN102118433A (en) | Multiple-tier distributed cluster system | |
| CN101150485A (en) | A management method for network data transmission of zero copy buffer queue | |
| WO2019127018A1 (en) | Memory system access method and device | |
| CN112948149A (en) | Remote memory sharing method and device, electronic equipment and storage medium | |
| CN106020731B (en) | Store equipment, array of storage devices and network adapter | |
| CN103517085B (en) | Method for implementing remote server management based on video decoding design | |
| CN112199309A (en) | Data reading method and device based on DMA engine and data transmission system | |
| CN110199512B (en) | Management method and device for storage equipment in storage system | |
| CN114629849A (en) | Storage device and operation method thereof | |
| US20090198699A1 (en) | Remote space efficient repository | |
| WO2023098050A1 (en) | Remote data access method and apparatus | |
| CN100459700C (en) | Method for realizing video requesting system based on protocol buffer storage | |
| Hines et al. | Distributed anemone: Transparent low-latency access to remote memory | |
| WO2023246843A1 (en) | Data processing method, apparatus and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |