CN103647708A - ATCA-based data message processing board - Google Patents
ATCA-based data message processing boardDownload PDFInfo
- Publication number
- CN103647708A CN103647708ACN201310631459.7ACN201310631459ACN103647708ACN 103647708 ACN103647708 ACN 103647708ACN 201310631459 ACN201310631459 ACN 201310631459ACN 103647708 ACN103647708 ACN 103647708A
- Authority
- CN
- China
- Prior art keywords
- information
- message
- data
- header information
- data message
- 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
- JEOQACOXAOEPLX-WCCKRBBISA-N(2s)-2-amino-5-(diaminomethylideneamino)pentanoic acid;1,3-thiazolidine-4-carboxylic acidChemical compoundOC(=O)C1CSCN1.OC(=O)[C@@H](N)CCCN=C(N)NJEOQACOXAOEPLX-WCCKRBBISA-N0.000titleclaimsabstractdescription23
- 238000012545processingMethods0.000titleclaimsabstractdescription19
- 230000015654memoryEffects0.000claimsdescription16
- 238000012546transferMethods0.000claimsdescription15
- 238000000034methodMethods0.000description5
- 230000005540biological transmissionEffects0.000description4
- 238000004891communicationMethods0.000description4
- 230000006870functionEffects0.000description4
- NBGBEUITCPENLJ-UHFFFAOYSA-NBunazosin hydrochlorideChemical compoundCl.C1CN(C(=O)CCC)CCCN1C1=NC(N)=C(C=C(OC)C(OC)=C2)C2=N1NBGBEUITCPENLJ-UHFFFAOYSA-N0.000description2
- 238000010586diagramMethods0.000description2
- 230000003287optical effectEffects0.000description2
- 238000005303weighingMethods0.000description2
- 230000009471actionEffects0.000description1
- 230000000694effectsEffects0.000description1
- 238000005516engineering processMethods0.000description1
- 239000004744fabricSubstances0.000description1
- 230000006872improvementEffects0.000description1
- 230000005012migrationEffects0.000description1
- 238000013508migrationMethods0.000description1
- 230000004048modificationEffects0.000description1
- 238000012986modificationMethods0.000description1
- 230000008569processEffects0.000description1
- 230000004044responseEffects0.000description1
- 230000009897systematic effectEffects0.000description1
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
Technical field
The present invention relates to communication engineering field, specifically, relate to a kind of based on ATCA(Advanced Telecom Computing Architecture, communication system computer platform framework) data message disposable plates.
Background technology
At present, bag disposable plates based on ATCA can support that the main flow implementation of 40G flow is to take two NP(Network Processor that REDSYS is representative, network processing unit) scheme namely has two completely independently NP on a plate, and each NP is responsible for the flow of 20G.Although realize simply like this, when practical application, also there is many weak points, for example:
The sometimes flow of two 20G unbalanced meeting causes a NP load overweight, if carry out dynamic load migration, can produce extra expense again; In addition, because some algorithm of customer requirement is not suitable for realizing with NP, can seriously increase the pressure of NP access memory, cause the workload of software tuning to increase severely; In addition, two NP schemes are difficult to directly access the flow of 40G on plank, need an extra pre-processed board or interface board, have reduced like this bulk density of whole system in the situation that ATCA standard has been stipulated cabinet number of slots; In addition, two NP schemes generally can adopt an exchange chip to do exchange in plate, to make up the problem of NP network interface lazy weight and outer net expansion interface.But exchange chip cost very golf calorific value brings very large problem to system very greatly, and increase the transmission delay that has significantly increased system after one-level exchanges, be unfavorable for the performance of systematic function.
The mass flow discrepancy that can support the main flow implementation of 40G flow to exist for the bag disposable plates based on ATCA in existing correlation technique weighs, internal storage access pressure is excessive, need to configure extra pre-processed board or interface board and need to configure extra exchange chip does the problem exchanging in class, not yet proposes at present effective solution.
Summary of the invention
For the bag disposable plates based on ATCA in existing correlation technique, can support that the mass flow discrepancy of the main flow implementation existence of 40G flow weighs, internal storage access pressure is excessive, need to configure extra pre-processed board or interface board and need to configure extra exchange chip and do the problem exchanging in plate, the present invention proposes a kind of data message disposable plates based on ATCA.
Technical scheme of the present invention is achieved in that
According to an aspect of the present invention, provide a kind of data message disposable plates based on ATCA.
Should the data message disposable plates based on ATCA comprise:
FPGA, for receiving data packets, and analyzes the described data message receiving, and determines the header information that described data message is corresponding, and sends this header information;
Coprocessor, be used for receiving described header information, and according to pre-configured header processing rule, determine the message executable operations information suitable with described header information, and send this message executable operations information, wherein, described header processing rule for record header information and with the corresponding relation of the suitable message executable operations information of this header information;
Described FPGA is also for receiving described message executable operations information, and according to operation corresponding to this message executable operations information and executing.
Wherein, header information comprise following one of at least: the source port information of the source address information of data transfer layer protocol information, transfer of data, the destination address information of transfer of data, transfer of data, the destination interface information of transfer of data.
In addition, should the data message disposable plates based on ATCA also comprise: first memory, for determining after described header information at described FPGA, the data message that described FPGA is received carries out buffer memory.
In addition, described FPGA also, for according to the corresponding relation of pre-configured header information and coprocessor core information, determines the coprocessor core information that described header information is corresponding, and sends described coprocessor core information.
And, described coprocessor is also for receiving described coprocessor core information, and according to this coprocessor core information, described header information is dispensed to corresponding coprocessor core, impel described coprocessor core according to pre-configured header processing rule, determine the message executable operations information suitable with described header information.
In addition, should the data message disposable plates based on ATCA also comprise: second memory, for storing the corresponding relation of pre-configured header information and coprocessor core information.
The data traffic of the data message that wherein, described FPGA receives is 40G.
The present invention is by adopting the direct receiving data packets of FPGA, thereby avoided disposable plates when receiving data packets, to occur that the phenomenon of mass flow discrepancy weighing apparatus occurs, also avoided needing the trouble of additional configuration pre-processed board or interface board simultaneously, in addition, due to the processor that has adopted programmable FPGA as disposable plates, thereby make the algorithm that user requires that meets that data message disposable plates can adaptivity, avoid the excessive phenomenon of internal storage access pressure to occur, and because FPGA possesses data exchanging function in andante, from having avoided needing additional configuration exchange chip to do the trouble exchanging in plate, reduced the input cost of equipment, guaranteed the transmission rate of system, improved the performance performance of system.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is according to the structural representation of the data message disposable plates based on ATCA of the embodiment of the present invention;
Fig. 2 is according to the principle configuration diagram of the data message disposable plates based on ATCA of the embodiment of the present invention;
Fig. 3 is the principle schematic while processing according to the data flow of the data message disposable plates based on ATCA of the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, the every other embodiment that those of ordinary skills obtain, belongs to the scope of protection of the invention.
According to embodiments of the invention, provide a kind of data message disposable plates based on ATCA.
As shown in Figure 1, according to the data message disposable plates based on ATCA of the embodiment of the present invention, comprise:
FPGA11, for receiving data packets, and analyzes the described data message receiving, and determines the header information that described data message is corresponding, and sends this header information;
Described FPGA11 is also for receiving described message executable operations information, and according to operation corresponding to this message executable operations information and executing.
Wherein, header information comprise following one of at least: the source port information of the source address information of data transfer layer protocol information, transfer of data, the destination address information of transfer of data, transfer of data, the destination interface information of transfer of data.
In addition, should the data message disposable plates based on ATCA also comprise: first memory (not shown), for determining after described header information at described FPGA, the data message that described FPGA is received carries out buffer memory.
In addition, described FPGA11 also, for according to the corresponding relation of pre-configured header information and coprocessor core information, determines the coprocessor core information that described header information is corresponding, and sends described coprocessor core information.
And, describedcoprocessor 12 is also for receiving described coprocessor core information, and according to this coprocessor core information, described header information is dispensed to corresponding coprocessor core, impel described coprocessor core according to pre-configured header processing rule, determine the message executable operations information suitable with described header information.
In addition, should the data message disposable plates based on ATCA also comprise: second memory (not shown), for storing the corresponding relation of pre-configured header information and coprocessor core information.
The data traffic of the data message that wherein, described FPGA11 receives is 40G.
From concrete principle aspect, technique scheme of the present invention is elaborated below.
Fig. 2 is the principle configuration diagram of the data message disposable plates based on ATCA, as can be seen from Figure 2,, in whole framework, FPGA is the control centre of flow, played the effect of master chip, coprocessor (XLP) is that the companion chip as FPGA is used.Below respectively FPGA part and XLP are partly described.
To with FPGA part, in whole framework, FPGA has the SFI5.1 interface of a 17*3.125G to be connected on the 300PIN optical module of a 40G, in order to accept the input flow rate of single channel 40G, (be mainly 40GPOS, compatible 40GE), certainly, also can be connected to one with the subcard of 4 XFP modules, in order to realize the access of 4*10GPOS.
And in order to improve throughput of system, it is second memory that FPGA has also connected two TCAM() be used for pre-search rule, between two TCAM, adopted the mode of serial connection to connect the RX that the TX of FPGA is namely connected to TCAM1, the TX of TCAM1 is connected to the RX of TCAM2, the TX of TCAM2 is connected to the RX of FPGA, is all 12*6.25GILA bus.And, on FGPA, also connected 4 QDR chips (being first memory), wherein, the space of each QDR chip is 72Mb, be used for data cached stream when XLP searches in detail, each QDR is used independently controller to widen system cache total bandwidth.
In addition, in order to realize the communication of data, FPGA is upper also has the IL signal of one group of 40G bandwidth to be connected to xlp, and for the communication of self-defined association processing messages, indication xlp completes the operation of response, and the lookup result that obtains returning from XLP.And, FPGA has also realized the function of a part of exchange chip, concrete, FPGA has 2 40GE network interfaces to be connected to Zone2 connector, the service management data that are used for CPB to send are transmitted to XLP by 10,000,000,000 interfaces between FPGA and XLP, a part of flow can also be given to other PPB simultaneously to realize load balancing.Because FPGA does not support the electric agreement of 40GE-KR4, so be connected in series a BCM84328 on link, the QSFP+ bus of FPGA is changed into KR4 standard.In addition, FPGA has 16 10GE and 24 interfaces of 8 GE totals to be connected to zone3 connector, coordinates special a RTM board to realize descending being redirected or backflow interface.
And for XLP part, in whole framework, XLP is that the primary processor of writing processor and system management as data processing carrys out use.
Wherein, for data division, XIL has connected four DDR3 internal memories to preserve detailed rules data, when FPGA sends after searching message, in DDR3, searches.
Wherein, for administrative section, in XLP, need to have the control of photoelectric protection, need to communicate by letter with CPB with OPB, therefore need two GE networks to be connected to the base switching network of Zone2.In order to improve system business handling property, the administrative messag message of the rule list operation between CPB and PPB, walks the fabric exchange of 10GE, and the base exchange of GE is used for doing the management (being mainly hardware management) of independent of service specially, therefore, XLP has 10GE interface (XAUI) E to be connected on FPGA.And because the rule of FPGA is downloaded and the PCIE interface of managerial demand by XLP carries out, so XLP also to need to have PCIE to be connected to FPGA upper, the network interface of XLP and serial ports are drawn from front panel, facilitate system debug.Whether XLP has also connected SD card and the NAND Flash of startup use in addition, and by a circuit, automatically detect SD card and insert, if inserted, otherwise starts from SD card start-up from NAND Flash.
Fig. 3 is the principle schematic of the data flow of the data message disposable plates based on ATCA while processing, as can be seen from Figure 3, PPB is to each message of receiving, extract the five-tuple information (transport layer protocol of message, source ip, object ip, transport layer source port, transport layer destination interface), and search in a plurality of rule lists, which server decision is transmitted to by these five-tuple information is continued to analyze, or blocking-up or this message of continuation transmission, the present invention has used a FPGA and XLP on PPB, wherein, FPGA is responsible for flow access and buffer memory, data enter from the 40G optical module of front panel, extract afterwards the header information (comprising five-tuple information) of message, and according to header information, search the TCAM chip being connected with FPGA, and predict the regular hit situation of message in DDR3.Header information and prediction case, by interlacken interface, send to XLP, the rule list of preserving in XLP inquiry ddr3 internal memory, in this process, the initial data of message is kept in QDR buffer memory.XLP possesses 32 core cpus, wherein, the present invention uses No. 0 as CPU management, and the rule that other 31 CPU carry out service surface is tabled look-up, XLP receives the message five-tuple information transmitting from FPGA, by micro engine (MicroEngine) five-tuple distribution of information 31 CPU to service surface.Search after CPU completes rule searching in DDR3 result feedback sent back to FPGA to CPU0 unification,
FPGA determines the action that carry out message according to rule match result, have two kinds of mode of operations to be decided by the configuration of CPB here.
A kind of is shunt mode: if hit, to descending port repeat message otherwise dropping packets.E-Packet and likely have both direction, be specially between the plate of downlink port on RTM or Zone2 balancedly, FPGA determines from which mouthful to forward according to inner forwarding rule list and the result of hitting.All miss messages are all abandoned, and do not have any message and send from uplink port.
Another kind is filtered model: if hit, to descending port repeat message otherwise message is forwarded from another uplink port.To descending port repeat message, likely have both direction, be specially between the plate of downlink port on RTM or Zone2 balancedly, FPGA determines from which mouthful to forward according to inner forwarding rule list and the result of hitting.Whether no matter hit, most packet all will copy portion and send out from another uplink port, only has the packet that hits on a small quantity special rules only to descending port repeat, to there will not be on another uplink port.
In sum, by means of technique scheme of the present invention, by adopting the direct receiving data packets of FPGA, thereby avoided disposable plates when receiving data packets, to occur that the phenomenon of mass flow discrepancy weighing apparatus occurs, also avoided needing the trouble of additional configuration pre-processed board or interface board simultaneously, in addition, due to the processor that has adopted programmable FPGA as disposable plates, thereby make the algorithm that user requires that meets that data message disposable plates can adaptivity, avoid the excessive phenomenon of internal storage access pressure to occur, and because FPGA possesses data exchanging function in andante, from having avoided needing additional configuration exchange chip to do the trouble exchanging in plate, reduced the input cost of equipment, guaranteed the transmission rate of system, improved the performance performance of system.
The foregoing is only preferred embodiment of the present invention, in order to limit the present invention, within the spirit and principles in the present invention not all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (6)
1. the data message disposable plates based on Advanced telecom computing architecture ATCA, is characterized in that, comprising:
On-site programmable gate array FPGA, for receiving data packets, and analyzes the described data message receiving, and determines the header information that described data message is corresponding, and sends this header information;
Coprocessor, be used for receiving described header information, and according to pre-configured header processing rule, determine the message executable operations information suitable with described header information, and send this message executable operations information, wherein, described header processing rule for record header information and with the corresponding relation of the suitable message executable operations information of this header information;
Described FPGA is also for receiving described message executable operations information, and according to operation corresponding to this message executable operations information and executing.
2. data message disposable plates according to claim 1, is characterized in that, described header information comprise following one of at least:
The source port information of the source address information of data transfer layer protocol information, transfer of data, the destination address information of transfer of data, transfer of data, the destination interface information of transfer of data.
3. data message disposable plates according to claim 1, is characterized in that, further comprises:
First memory, for determining after described header information at described FPGA, the data message that described FPGA is received carries out buffer memory.
4. data message disposable plates according to claim 1, it is characterized in that, described FPGA also, for according to the corresponding relation of pre-configured header information and coprocessor core information, determines the coprocessor core information that described header information is corresponding, and sends described coprocessor core information;
And, described coprocessor is also for receiving described coprocessor core information, and according to this coprocessor core information, described header information is dispensed to corresponding coprocessor core, impel described coprocessor core according to pre-configured header processing rule, determine the message executable operations information suitable with described header information.
5. data message disposable plates according to claim 4, is characterized in that, further comprises:
Second memory, for storing the corresponding relation of pre-configured header information and coprocessor core information.
6. according to the data message disposable plates described in any one in claim 1 to 5, it is characterized in that, the data traffic of the data message that described FPGA receives is 40G.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310631459.7ACN103647708A (en) | 2013-11-29 | 2013-11-29 | ATCA-based data message processing board |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310631459.7ACN103647708A (en) | 2013-11-29 | 2013-11-29 | ATCA-based data message processing board |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103647708Atrue CN103647708A (en) | 2014-03-19 |
Family
ID=50252868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310631459.7APendingCN103647708A (en) | 2013-11-29 | 2013-11-29 | ATCA-based data message processing board |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103647708A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106789706A (en)* | 2016-11-11 | 2017-05-31 | 天津光电通信技术有限公司 | A Network Distribution System Based on TCAM |
| CN111404798A (en)* | 2020-03-09 | 2020-07-10 | 湖北微源卓越科技有限公司 | System and method for multi-user rule matching and flow replication |
| CN114124822A (en)* | 2021-11-29 | 2022-03-01 | 杭州迪普信息技术有限公司 | Message matching processing device and method |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1545254A (en)* | 2003-11-13 | 2004-11-10 | 中兴通讯股份有限公司 | A method of fast data packet filtering |
| CN1642146A (en)* | 2005-01-07 | 2005-07-20 | 清华大学 | Bag-preprocessing circuit assembly of interface card for high-speed network diversion equipment |
| CN1674557A (en)* | 2005-04-01 | 2005-09-28 | 清华大学 | Parallel IP packet sorter matched with settling range based on TCAM and method thereof |
| US20050281267A1 (en)* | 2004-06-22 | 2005-12-22 | Mcfadden Gordon | ATM switching protocol |
| US20060140226A1 (en)* | 2004-12-28 | 2006-06-29 | Michael Ho | Techniques for processing traffic transmitted over advanced switching compatible switch fabrics |
| US20070038843A1 (en)* | 2005-08-15 | 2007-02-15 | Silicon Informatics | System and method for application acceleration using heterogeneous processors |
| CN1921457A (en)* | 2006-09-18 | 2007-02-28 | 杭州华为三康技术有限公司 | Network equipment and message transferring method based on multiple-core processor |
| CN101098293A (en)* | 2006-06-26 | 2008-01-02 | 西门子通信技术（北京）有限公司 | Universal packet wireless service network gateway supporting node device and fee counting method |
| CN101217467A (en)* | 2007-12-28 | 2008-07-09 | 杭州华三通信技术有限公司 | An inter-core load dispensing device and method |
| CN101217455A (en)* | 2007-01-05 | 2008-07-09 | 上海复旦光华信息科技股份有限公司 | A secure content filtering shunt based on the integration of useful connecting data |
| US20080244150A1 (en)* | 2007-02-02 | 2008-10-02 | Viswa Sharma | Processor chip arcitecture having integrated high-speed packet switched serial interface |
| US20080270399A1 (en)* | 2007-04-29 | 2008-10-30 | Bo Feng | Method and system for parallel flow-awared pattern matching |
| CN101599963A (en)* | 2009-06-10 | 2009-12-09 | 电子科技大学 | Network suspected threat information filter and screening processing method |
| CN101610209A (en)* | 2008-11-28 | 2009-12-23 | 北京网康科技有限公司 | A kind of Network method for stream processing and equipment of multi-core parallel concurrent |
| CN201388208Y (en)* | 2009-04-03 | 2010-01-20 | 南京贝伦思网络科技有限公司 | Multi-rate, multi-interface data message processing veneer based on ATCA |
| WO2010148054A2 (en)* | 2009-06-19 | 2010-12-23 | Singular Computing, Llc | Processing with compact arithmetic processing element |
| CN102004673A (en)* | 2010-11-29 | 2011-04-06 | 中兴通讯股份有限公司 | Processing method and system of multi-core processor load balancing |
| CN102710424A (en)* | 2012-05-30 | 2012-10-03 | 曙光信息产业(北京)有限公司 | Gigabit/10-gigabit multifunctional network card and implementation method for same |
| CN202750110U (en)* | 2012-09-07 | 2013-02-20 | 山东紫光比威网络技术有限公司 | High-end IPV6 safety dual-stack core switch |
| CN202856775U (en)* | 2012-02-15 | 2013-04-03 | 南京贝伦思网络科技有限公司 | Data message processing plate of 12 port 10-gigabit Ethernet and 40Gbps POS interface based on ATCT |
| CN103368777A (en)* | 2013-07-11 | 2013-10-23 | 曙光信息产业股份有限公司 | Data packet processing board and processing method |
- 2013
- 2013-11-29CNCN201310631459.7Apatent/CN103647708A/enactivePending
Patent Citations (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1545254A (en)* | 2003-11-13 | 2004-11-10 | 中兴通讯股份有限公司 | A method of fast data packet filtering |
| US20050281267A1 (en)* | 2004-06-22 | 2005-12-22 | Mcfadden Gordon | ATM switching protocol |
| US20060140226A1 (en)* | 2004-12-28 | 2006-06-29 | Michael Ho | Techniques for processing traffic transmitted over advanced switching compatible switch fabrics |
| CN1642146A (en)* | 2005-01-07 | 2005-07-20 | 清华大学 | Bag-preprocessing circuit assembly of interface card for high-speed network diversion equipment |
| CN1674557A (en)* | 2005-04-01 | 2005-09-28 | 清华大学 | Parallel IP packet sorter matched with settling range based on TCAM and method thereof |
| US20070038843A1 (en)* | 2005-08-15 | 2007-02-15 | Silicon Informatics | System and method for application acceleration using heterogeneous processors |
| CN101098293A (en)* | 2006-06-26 | 2008-01-02 | 西门子通信技术（北京）有限公司 | Universal packet wireless service network gateway supporting node device and fee counting method |
| CN1921457A (en)* | 2006-09-18 | 2007-02-28 | 杭州华为三康技术有限公司 | Network equipment and message transferring method based on multiple-core processor |
| CN101217455A (en)* | 2007-01-05 | 2008-07-09 | 上海复旦光华信息科技股份有限公司 | A secure content filtering shunt based on the integration of useful connecting data |
| US20080244150A1 (en)* | 2007-02-02 | 2008-10-02 | Viswa Sharma | Processor chip arcitecture having integrated high-speed packet switched serial interface |
| US20080270399A1 (en)* | 2007-04-29 | 2008-10-30 | Bo Feng | Method and system for parallel flow-awared pattern matching |
| CN101217467A (en)* | 2007-12-28 | 2008-07-09 | 杭州华三通信技术有限公司 | An inter-core load dispensing device and method |
| CN101610209A (en)* | 2008-11-28 | 2009-12-23 | 北京网康科技有限公司 | A kind of Network method for stream processing and equipment of multi-core parallel concurrent |
| CN201388208Y (en)* | 2009-04-03 | 2010-01-20 | 南京贝伦思网络科技有限公司 | Multi-rate, multi-interface data message processing veneer based on ATCA |
| CN101599963A (en)* | 2009-06-10 | 2009-12-09 | 电子科技大学 | Network suspected threat information filter and screening processing method |
| WO2010148054A2 (en)* | 2009-06-19 | 2010-12-23 | Singular Computing, Llc | Processing with compact arithmetic processing element |
| WO2010148054A3 (en)* | 2009-06-19 | 2011-03-31 | Singular Computing, Llc | Processing with compact arithmetic processing element |
| CN102004673A (en)* | 2010-11-29 | 2011-04-06 | 中兴通讯股份有限公司 | Processing method and system of multi-core processor load balancing |
| CN202856775U (en)* | 2012-02-15 | 2013-04-03 | 南京贝伦思网络科技有限公司 | Data message processing plate of 12 port 10-gigabit Ethernet and 40Gbps POS interface based on ATCT |
| CN102710424A (en)* | 2012-05-30 | 2012-10-03 | 曙光信息产业(北京)有限公司 | Gigabit/10-gigabit multifunctional network card and implementation method for same |
| CN202750110U (en)* | 2012-09-07 | 2013-02-20 | 山东紫光比威网络技术有限公司 | High-end IPV6 safety dual-stack core switch |
| CN103368777A (en)* | 2013-07-11 | 2013-10-23 | 曙光信息产业股份有限公司 | Data packet processing board and processing method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106789706A (en)* | 2016-11-11 | 2017-05-31 | 天津光电通信技术有限公司 | A Network Distribution System Based on TCAM |
| CN106789706B (en)* | 2016-11-11 | 2020-08-07 | 天津光电通信技术有限公司 | Network shunting system based on TCAM |
| CN111404798A (en)* | 2020-03-09 | 2020-07-10 | 湖北微源卓越科技有限公司 | System and method for multi-user rule matching and flow replication |
| CN114124822A (en)* | 2021-11-29 | 2022-03-01 | 杭州迪普信息技术有限公司 | Message matching processing device and method |
| CN114124822B (en)* | 2021-11-29 | 2024-04-26 | 杭州迪普信息技术有限公司 | Message matching processing device and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gangidi et al. | Rdma over ethernet for distributed training at meta scale | |
| Derradji et al. | The BXI interconnect architecture | |
| US9563735B1 (en) | Automatic pipelining of NoC channels to meet timing and/or performance | |
| US9699079B2 (en) | Streaming bridge design with host interfaces and network on chip (NoC) layers | |
| US7865654B2 (en) | Programmable bridge header structures | |
| US11372787B2 (en) | Unified address space for multiple links | |
| US7830882B2 (en) | Switch scaling for virtualized network interface controllers | |
| US9647962B2 (en) | Non-transparent bridge method and apparatus for configuring high-dimensional PCI-express networks | |
| US20190188178A1 (en) | Multiple transaction data flow control unit for high-speed interconnect | |
| EP4016309A1 (en) | System, apparatus and method for handling multi-protocol traffic in data link layer circuitry | |
| CN101242371A (en) | Method, system and device router overlapping based on PCIe exchange architecture | |
| CN103368777B (en) | A kind of processing data packets plate and processing method | |
| Shu et al. | Direct Universal Access: Making Data Center Resources Available to {FPGA} | |
| US7643477B2 (en) | Buffering data packets according to multiple flow control schemes | |
| CN104247341A (en) | Dynamic optimization of a multicast tree hierarchy for a distributed switch | |
| US20070127365A1 (en) | Method for implementing redundant structure of ATCA (advanced telecom computing architecture) system via base interface and the ATCA system for use in the same | |
| CN103647708A (en) | ATCA-based data message processing board | |
| US20250267065A1 (en) | In-switch collective primitive processing | |
| Saljoghei et al. | dreddbox: Demonstrating disaggregated memory in an optical data centre | |
| WO2013162569A1 (en) | Increasing a data transfer rate | |
| CN106534400A (en) | Network address conversion method and device | |
| Bu-khamsin | Socket direct protocol over PCI express interconnect: design, implementation and evaluation | |
| US10084725B2 (en) | Extracting features from a NoC for machine learning construction | |
| Zang et al. | PROP: Using PCIe-based RDMA to accelerate rack-scale communications in data centers | |
| CN105743819B (en) | Computing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20140319 |