CN111124967A - Expansion system capable of locally strengthening interconnection based on high-throughput computing platform - Google Patents
Expansion system capable of locally strengthening interconnection based on high-throughput computing platform Download PDFInfo
- Publication number
- CN111124967A CN111124967A CN201911142391.XA CN201911142391A CN111124967A CN 111124967 A CN111124967 A CN 111124967A CN 201911142391 A CN201911142391 A CN 201911142391A CN 111124967 A CN111124967 A CN 111124967A
- Authority
- CN
- China
- Prior art keywords
- unit
- platform
- cache
- direct memory
- central processing
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/382—Information transfer, e.g. on bus using universal interface adapter
- G06F13/385—Information transfer, e.g. on bus using universal interface adapter for adaptation of a particular data processing system to different peripheral devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/08—Protocols for interworking; Protocol conversion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2213/00—Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F2213/0026—PCI express
Abstract
The invention discloses an expansion system capable of locally strengthening interconnection based on a high-throughput computing platform, and relates to the technical field of high-throughput computing. The system comprises a central processing unit, a storage platform, an expansion interface unit, a protocol control unit, an accelerated calculation unit group, an IO platform, a cache/direct memory unit, an Ethernet interface and a switch, wherein the accelerated calculation unit group, the IO platform and the cache/direct memory unit are respectively connected with the central processing unit through the expansion interface unit and the Ethernet interface, and the protocol control unit is arranged among the accelerated calculation unit group, the IO platform, the cache/direct memory unit and the expansion interface unit. The invention takes the switch as an extension foundation, is connected with more accelerating computing components through the extension of the switch, utilizes the Ethernet interface to count the computing data, controls each unit through the extension interface unit, can carry out local reinforced extension on the functions of the high-throughput computing platform, and further improves the computing capability of the high-throughput computing platform.
Description
Technical Field
The invention belongs to the technical field of high-throughput computing, and particularly relates to an expansion system capable of locally strengthening interconnection based on a high-throughput computing platform.
Background
With the advance of biochemical engineering, the complexity of materials is higher and higher, the reactant process is also more and more complex, the calculation time can be effectively reduced by introducing high-throughput calculation, and the calculation process is more reliable. At present, when a high-throughput computing platform is used for computing, due to different purposes, functions and computing power of the high-throughput computing platform are often fixed, certain functions cannot be strengthened, the practicability is reduced in complex and changeable experiments, the cost of the high-throughput computing platform is high, and obviously, the cost of equipment with higher replacement computing power is overlarge.
Disclosure of Invention
The invention aims to provide an expansion system capable of locally strengthening interconnection based on a high-throughput computing platform, and solves the problem that the high-throughput computing platform is difficult to expand in the background art.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the expansion system based on the high-throughput computing platform and capable of locally strengthening interconnection comprises a central processing unit, a storage platform, an expansion interface unit, a protocol control unit, an accelerated computing unit group, an IO platform, a cache/direct memory unit, an Ethernet interface and a switch;
the switch is respectively connected with the acceleration unit group, the IO platform and the cache/direct memory unit, the switch interacts data of the acceleration unit group, the IO platform and the cache/direct memory unit, the acceleration calculation unit group, the IO platform and the cache/direct memory unit are all connected with the central processing unit through Ethernet interfaces, the rapid calculation unit group, the IO platform and the cache/direct memory unit send device information to the central processing unit through the Ethernet interfaces, the acceleration calculation unit group is respectively connected with the storage platform and the central processing unit through the protocol control unit and the expansion interface unit in sequence, the central processing unit respectively controls the acceleration calculation unit group, the IO platform and the cache/direct memory unit through the expansion interface unit and the protocol control unit, meanwhile, the central processing unit stores the operation data and the calculation data of the equipment in the storage platform through the interface unit.
Preferably, the expansion interface unit is an interface supporting PCIe protocol.
Preferably, the acceleration computing unit group comprises a plurality of acceleration computing units, and the acceleration computing units are connected in parallel.
Preferably, the storage platform and the expansion interface unit interact data in a parallel transmission mode.
Preferably, the ethernet interface is an interface supporting gigabit ethernet protocol.
The invention has the following beneficial effects:
the invention takes the switch as an extension foundation, is connected with more accelerating computing components through the extension of the switch, utilizes the Ethernet interface to count the computing data, controls each unit through the extension interface unit, can carry out local reinforced extension on the functions of the high-throughput computing platform, and further improves the computing capability of the high-throughput computing platform.
Of course, it is not necessary for any product in which the invention is practiced to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is an expanded system block diagram of the high-throughput computing platform based interconnect that can be locally enhanced according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the expansion system capable of locally strengthening interconnection based on a high-throughput computing platform includes a central processing unit, a storage platform, an expansion interface unit, a protocol control unit, an accelerated computing unit group, an IO platform, a cache/direct memory unit, an ethernet interface, and a switch;
the switch is respectively connected with the acceleration unit group, the IO platform and the cache/direct memory unit, the switch interacts data of the acceleration unit group, the IO platform and the cache/direct memory unit, the acceleration calculation unit group, the IO platform and the cache/direct memory unit are all connected with the central processing unit through Ethernet interfaces, the rapid calculation unit group, the IO platform and the cache/direct memory unit send equipment information to the central processing unit through the Ethernet interfaces, the acceleration calculation unit group is respectively connected with the storage platform and the central processing unit through the protocol control unit and the expansion interface unit in sequence, the central processing unit respectively controls the acceleration calculation unit group, the IO platform and the cache/direct memory unit through the expansion interface unit and the protocol control unit, meanwhile, the central processing unit stores the operation data and the calculation data of the equipment in the storage platform through the interface unit.
The expansion interface unit is an interface supporting PCIe protocol.
The acceleration computing unit group comprises a plurality of acceleration computing units, and the acceleration computing units are connected in parallel.
The storage platform and the expansion interface unit interact data in a parallel transmission mode.
The Ethernet interface is an interface supporting gigabit Ethernet protocol.
Referring to fig. 1, the present invention is an expansion system capable of local enhanced interconnection based on a high throughput computing platform, the control principle is that parameter input is performed through an IO platform, an accelerated computing unit performs computing processing on the input parameters, the input parameters and data results processed by the accelerated computing unit are temporarily stored in a cache/direct memory unit to facilitate data calling, and simultaneously the data processing information is transmitted to a central processing unit through an ethernet interface, the central processing unit controls each unit through an expansion interface unit, the accelerated computing unit, the IO platform and the cache/direct memory unit are respectively connected with the expansion interface unit through different control protocols, the central processing unit identifies each unit through identifying different control protocols and controls each unit, and the central processing unit stores the control information, the computing results of the accelerated computing unit and the input parameter data in a storage platform, convenient to look up and call.
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (5)
1. The expansion system based on the high-throughput computing platform and capable of locally strengthening interconnection comprises a central processing unit, a storage platform, an expansion interface unit, a protocol control unit, an accelerated computing unit group, an IO platform, a cache/direct memory unit, an Ethernet interface and a switch, and is characterized in that:
the switch is respectively connected with the acceleration unit group, the IO platform and the cache/direct memory unit, the switch interacts data of the acceleration unit group, the IO platform and the cache/direct memory unit, the acceleration calculation unit group, the IO platform and the cache/direct memory unit are all connected with the central processing unit through Ethernet interfaces, the rapid calculation unit group, the IO platform and the cache/direct memory unit send device information to the central processing unit through the Ethernet interfaces, the acceleration calculation unit group is respectively connected with the storage platform and the central processing unit through the protocol control unit and the expansion interface unit in sequence, the central processing unit respectively controls the acceleration calculation unit group, the IO platform and the cache/direct memory unit through the expansion interface unit and the protocol control unit, meanwhile, the central processing unit stores the operation data and the calculation data of the equipment in the storage platform through the interface unit.
2. The high-throughput computing platform-based expansion system capable of locally enforcing interconnections according to claim 1, wherein the expansion interface unit is an interface supporting PCIe protocol.
3. The high-throughput computing platform-based expansion system capable of locally enforcing interconnection according to claim 1, wherein the set of accelerated computing units comprises a plurality of accelerated computing units, and the plurality of accelerated computing units are connected in parallel.
4. The high-throughput computing platform-based expansion system capable of locally strengthening interconnection is characterized in that data is interacted between the storage platform and the expansion interface unit in a parallel transmission mode.
5. The high-throughput computing platform-based expansion system capable of locally enforcing interconnection according to claim 1, wherein the ethernet interface is an interface supporting gigabit ethernet protocol.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911142391.XA CN111124967A (en) | 2019-11-20 | 2019-11-20 | Expansion system capable of locally strengthening interconnection based on high-throughput computing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911142391.XA CN111124967A (en) | 2019-11-20 | 2019-11-20 | Expansion system capable of locally strengthening interconnection based on high-throughput computing platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111124967A true CN111124967A (en) | 2020-05-08 |
Family
ID=70495891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911142391.XA Pending CN111124967A (en) | 2019-11-20 | 2019-11-20 | Expansion system capable of locally strengthening interconnection based on high-throughput computing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111124967A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5961627A (en) * | 1997-06-18 | 1999-10-05 | Gigalabs, Inc | High speed switch for high speed connection between computers and peripheral devices |
| CN101354694A (en) * | 2007-07-26 | 2009-01-28 | 上海红神信息技术有限公司 | Ultra-high expanding super computing system based on MPU structure |
| CN103778100A (en) * | 2012-10-23 | 2014-05-07 | 亚德诺半导体技术公司 | Memory interconnect network architecture for vector processor |
| CN104657297A (en) * | 2015-02-03 | 2015-05-27 | 杭州士兰控股有限公司 | Computing equipment expanding system and expanding method |
| CN204833253U (en) * | 2015-08-21 | 2015-12-02 | 北京盛讯美恒科技发展有限公司 | A USB server for cloud calculates |
| CN105191140A (en) * | 2013-03-15 | 2015-12-23 | 加利福尼亚大学董事会 | Network architectures for boundary-less hierarchical interconnects |
| CN106844263A (en) * | 2016-12-26 | 2017-06-13 | 中国科学院计算技术研究所 | It is a kind of based on configurable multiprocessor computer system and implementation method |
| CN109844722A (en) * | 2016-08-12 | 2019-06-04 | 利奇得公司 | Breakdown fabric switch computing platform |
| CN110275850A (en) * | 2018-12-29 | 2019-09-24 | 中国科学院软件研究所 | The calculation method and device of space-based supercomputer platform |
-
2019
- 2019-11-20 CN CN201911142391.XA patent/CN111124967A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5961627A (en) * | 1997-06-18 | 1999-10-05 | Gigalabs, Inc | High speed switch for high speed connection between computers and peripheral devices |
| CN101354694A (en) * | 2007-07-26 | 2009-01-28 | 上海红神信息技术有限公司 | Ultra-high expanding super computing system based on MPU structure |
| CN103778100A (en) * | 2012-10-23 | 2014-05-07 | 亚德诺半导体技术公司 | Memory interconnect network architecture for vector processor |
| CN105191140A (en) * | 2013-03-15 | 2015-12-23 | 加利福尼亚大学董事会 | Network architectures for boundary-less hierarchical interconnects |
| CN104657297A (en) * | 2015-02-03 | 2015-05-27 | 杭州士兰控股有限公司 | Computing equipment expanding system and expanding method |
| CN204833253U (en) * | 2015-08-21 | 2015-12-02 | 北京盛讯美恒科技发展有限公司 | A USB server for cloud calculates |
| CN109844722A (en) * | 2016-08-12 | 2019-06-04 | 利奇得公司 | Breakdown fabric switch computing platform |
| CN106844263A (en) * | 2016-12-26 | 2017-06-13 | 中国科学院计算技术研究所 | It is a kind of based on configurable multiprocessor computer system and implementation method |
| CN110275850A (en) * | 2018-12-29 | 2019-09-24 | 中国科学院软件研究所 | The calculation method and device of space-based supercomputer platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103166811A (en) | Multi-active detection (MAD) detection method and device | |
| CN105516292A (en) | Hot standby method of cloud platform of intelligent substation | |
| CN111193971B (en) | Machine learning-oriented distributed computing interconnection network system and communication method | |
| CN106598888B (en) | A kind of more board communication systems and method using RS485 agreement | |
| CN103092735A (en) | Method for updating node states | |
| CN111124967A (en) | Expansion system capable of locally strengthening interconnection based on high-throughput computing platform | |
| CN109714124B (en) | Time synchronization system and method for test system | |
| CN111475368A (en) | Serial port cascade regulation and control method and serial port equipment | |
| CN106054671A (en) | Intelligent vehicle simulation platform for active safety | |
| CN101916206A (en) | Method and device for modeling signal system | |
| CN104123261B (en) | A kind of electronic equipment and information transferring method | |
| CN114915516A (en) | Communication method and device | |
| CN203012478U (en) | Distributed control facility for ultra-multichannel optical fiber system of laser device | |
| CN114039810A (en) | Flexible automatic control system based on Ethernet | |
| CN110098989A (en) | A kind of multichannel CAN analogue system and test method based on CANFD bus | |
| CN104656478B (en) | The control circuit and control method of a kind of multi-power module | |
| CN113325757A (en) | Industrial robot control system | |
| CN113179325A (en) | Multi-terminal collaborative interaction method and device, gateway box and medium | |
| CN105830368B (en) | Optical interconnection system, node, optical-fiber network controller and the method for transmitting data | |
| CN113138732A (en) | Multi-display controller system, access control system, and display controller | |
| CN105353649B (en) | Direct current transportation analogue system and emulation mode based on MMC | |
| Kutuzov | The structure and modeling results of the parallel spatial switching system | |
| CN101478427A (en) | Method for accelerating PBT tunnel failure fast switching | |
| CN107276845A (en) | A kind of real time data dispatching method with fault tolerance of field of track traffic | |
| CN115643558B (en) | Data processing method and device, electronic equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200508 |
|
| RJ01 | Rejection of invention patent application after publication |