CN203243359U - Novel high-performance network data processing application server - Google Patents
Novel high-performance network data processing application server Download PDFInfo
- Publication number
- CN203243359U CN203243359U CN201320257369.1U CN201320257369U CN203243359U CN 203243359 U CN203243359 U CN 203243359U CN 201320257369 U CN201320257369 U CN 201320257369U CN 203243359 U CN203243359 U CN 203243359U
- Authority
- CN
- China
- Prior art keywords
- module
- network data
- npu
- chip
- application server
- 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.)
- Expired - Fee Related
Links
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 238000000034 method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000007906 compression Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000002708 enhancing Effects 0.000 description 2
- 238000011030 bottleneck Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Abstract
The utility model discloses a novel high-performance network data processing application server. The server comprises a CPU module, a system hub chip, a memory module, an input-output controller, an NPU module and extended communication ports. The CPU module is connected to the system hub chip; the system hub chip is connected to the memory module and the input-output controller which is connected to the NPU module through a PCI-E channel; the NPU module is connected to the extended communication ports; and the CPU module carries out business cooperation through the system hub chip and an NPU which uses a special-purpose MIPS architecture. Network data processing with the NPU, which employs a special-purpose MIPS architecture, rather than with a conventional X86 CPU, substantially improves the capabilities of network data processing, inputting and outputting. As a result, the production efficiency is largely improved, the cost invested on production equipment is reduced, and the server is more advantageous in deployment and better in energy saving and environmental protection.
Description
Technical field:
The utility model relates to a kind of novel high-performance network data processing application server, is particularly useful for the network message data analysis application.
Background technology:
The fast development of search engine, cloud computing, the Internet retail, Web2.0, mobile device, online game etc. makes Internet traffic reach unprecedented height, usually needs a large amount of server zones and 10Gbps, 20Gbps, 40Gbps even the higher network bandwidth.Simultaneously along with the variation of network application and complicated, the perception demand of using is become new focus, deep-packet detection and safe handling etc. require constantly to increase, and the demand that causes network infrastructure to process the bandwidth aspect is doubled and redoubled.Present 80% user network center is take the x86 framework as main, efficient is generally lower, performance loss is very large, in application, the load capacity of CPU is often less than 20%, bottleneck is present in the computing of network data and X86-based and exists huge delay and the transfer problem of instruction set, causes the large and waste of cost.In order to reach business demand, the corollary equipment of traditional server can produce huge renewal cost; On safeguarding, because the relation of quantity also can produce imponderable stack cost, affect the cost that datamation is analyzed.
Summary of the invention:
In order to solve input and output rate the value lower and with high costs deficiency of traditional server in the massive dataflow processing is used, the utility model provides a kind of novel high-performance network data processing application server, not only can greatly enhance productivity, reduce the Meteorological of production equipment, and have better deployable.
In order to solve the existing problem of background technology, the utility model is by the following technical solutions: a kind of novel high-performance network data processing application server, comprise the CPU module, the system backbone chip, memory modules, IOC, the NPU module, the expanding communication port, the CPU module links to each other with the system backbone chip, the system backbone chip links to each other with memory modules, the system backbone chip links to each other with IOC, IOC links to each other with the NPU module by the PCI-E channel, the NPU module links to each other with the expanding communication port, and the CPU module is carried out business collaboration by the NPU of the MIPS framework of system backbone chip and special use;
Wherein, described CPU module is selected Intel E52600 family chip; Described system backbone chip is Intel C602 chipset; Described memory modules is comprised of 18 DDR3 memory chips; Described NPU module adopts Cavium CN6XX series MIPS chip; Described expanding communication port is 2 10GE/POS communication interfaces.
The utility model contrast prior art has following beneficial effect: this practical high-performance network data is processed application server, the NPU of the mips framework by special use processes the network data that traditional X-ray 86CPU processes, greatly put forward network data processing according to handling capacity, thereby not only can greatly enhance productivity, reduce the Meteorological of production equipment, and have better deployment advantage, and more energy-conserving and environment-protective.
Description of drawings:
Fig. 1 is the utility model structural representation.
Embodiment:
Referring to Fig. 1, the utility model provides a kind of novel high-performance network data processing application server, comprise CPU module (1), system backbone chip (2), memory modules (3), IOC (4), NPU module (5), expanding communication port (6), CPU module (1) links to each other with system backbone chip (2), system backbone chip (2) links to each other with memory modules (3), system backbone chip (2) links to each other with IOC (4), IOC (4) links to each other with NPU module (5) by the PCI-E channel, NPU module (5) links to each other with expanding communication port (6), and CPU module (1) is carried out business collaboration by the NPU (5) of the MIPS framework of system backbone chip (2) and special use.
As preferably, described CPU module (1) is selected Intel E52600 family chip; Described system backbone chip (2) is Intel C602 chipset; Described memory modules (3) is comprised of 18 DDR3 memory chips; Described NPU module (5) adopts Cavium CN6XX series MIPS chip; Described expanding communication port (6) is 2 10GE/POS communication interfaces.
When work, the utility model is based on mips architecture network process chip standard, bundling technology by polymerization aspect the server network I/O medium, adopt 2 10Ge/pos expanding communication interfaces, can in the support that need not the CPU module, directly process the message hardware deep message and detect (DPI), the data message load balancing, Packet Classification.After the special-purpose NPU of the mips framework on the above-mentioned functions processed, the conventional parameter such as regular complexity and hit rate and regular number and CPU module (1) did not produce any performance load, greatly Optimized Service device overall performance.It is constant that network flow 10gbps fluxion reaches 10 retentions, and the message byte is on average reported when reaching 512 byte can bring into play optimum performance.The utility model is supported based on windows and linux kernel, can realize by the standard interface of kernel the kernel internal memory of management NPU module (5), realize (Dynamic Memory Allocation), CPU module (1) can be according to the standard interface of os system kernel, generate the bordereau of desired data, based on based on model to data analysis, and the network data processing function such as configuration management intelligent data network interface card.
Aspect the network data message forwarding performance, high-performance network data described in the utility model is processed the mips framework NPU module (5) of application server and is supported three layers of parcel literary composition transfer capability to 20Gbps, transfer capability that can be from the physics end to Linux user's attitude 13.5Gbps (average message length 1024 bytes), aspect data compression, support zlib (RFC1950), deflate (RFC1951), gzip (RFC1952), the following international agreement standard of LZS (RFC2395).Through test, direct network data compression all reaches about 3 times with the performance gap that CPU carries out the software processing based on mips, and reasonable level is down in the collaborative load of CPU.
Claims (6)
1. novel high-performance network data processing application server, it is characterized in that, comprise CPU module (1), system backbone chip (2), memory modules (3), IOC (4), NPU module (5), expanding communication port (6), CPU module (1) links to each other with system backbone chip (2), system backbone chip (2) links to each other with memory modules (3), system backbone chip (2) links to each other with IOC (4), IOC (4) links to each other with NPU module (5) by the PCI-E channel, NPU module (5) links to each other with expanding communication port (6), and CPU module (1) is carried out business collaboration by the NPU (5) of the MIPS framework of system backbone chip (2) and special use.
2. described high-performance network data according to claim 1 is processed application server, it is characterized in that, described CPU module (1) is selected Intel E52600 family chip.
3. high-performance network data according to claim 1 is processed application server, it is characterized in that, described system backbone chip (2) is Intel C602 chipset.
4. high-performance network data according to claim 1 is processed application server, it is characterized in that, described memory modules (3) is comprised of 18 DDR3 memory chips.
5. high-performance network data according to claim 1 is processed application server, it is characterized in that, described NPU module (5) adopts Cavium CN6XX series MIPS chip.
6. high-performance network data according to claim 1 is processed application server, it is characterized in that, described expanding communication port (6) is 2 10GE/POS communication interfaces.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201320257369.1U CN203243359U (en) | 2013-05-13 | 2013-05-13 | Novel high-performance network data processing application server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201320257369.1U CN203243359U (en) | 2013-05-13 | 2013-05-13 | Novel high-performance network data processing application server |
Publications (1)
Publication Number | Publication Date |
---|---|
CN203243359U true CN203243359U (en) | 2013-10-16 |
Family
ID=49320612
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201320257369.1U Expired - Fee Related CN203243359U (en) | 2013-05-13 | 2013-05-13 | Novel high-performance network data processing application server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN203243359U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105162657A (en) * | 2015-08-28 | 2015-12-16 | 浪潮电子信息产业股份有限公司 | Network testing performance optimization method |
-
2013
- 2013-05-13 CN CN201320257369.1U patent/CN203243359U/en not_active Expired - Fee Related
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105162657A (en) * | 2015-08-28 | 2015-12-16 | 浪潮电子信息产业股份有限公司 | Network testing performance optimization method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103034295B (en) | The reconfigurable micro server that I/O capability strengthens | |
CN105812346A (en) | Data interaction method of serial port device and Ethernet device | |
CN106790707A (en) | A kind of new cloud computing system | |
CN103916316A (en) | Linear speed capturing method of network data packages | |
CN203243359U (en) | Novel high-performance network data processing application server | |
CN204256732U (en) | The high-speed data transmission apparatus of Based PC I-Express interface | |
CN204229397U (en) | RS232 serial ports and ethernet interface converter | |
CN105573204A (en) | Multi-processor digital audio frequency matrix control device and method | |
US20180321882A1 (en) | Hardware system for data conversion and storage device | |
CN206021155U (en) | A kind of fusion architecture server | |
CN201878161U (en) | Host-to-host information security one-way transmission system | |
Xu et al. | Towards high-speed real-time http traffic analysis on the tilera many-core platform | |
CN104267911B (en) | Data storage controller and its data processing method | |
CN204360302U (en) | A kind of information management cloud stores all-in-one | |
CN106161430A (en) | application service device for data cluster | |
CN208013953U (en) | A kind of big data safe storage system | |
CN207543130U (en) | A kind of gateway system | |
CN202548713U (en) | Special computing node mainboard for cloud computing | |
CN205430271U (en) | Data processor | |
CN208890834U (en) | A kind of virtual network switching equipment of 4 based on DPDK network interface card times physical port | |
Wei et al. | Design of the web log analysis system based on hadoop | |
CN203911986U (en) | Cross-platform integrated data processing system | |
CN204667397U (en) | A kind of movement circuit being applicable to the flow collection equipment of networks congestion control information audit | |
CN203745830U (en) | Digital output controller used for Internet of things | |
CN203340107U (en) | Network transmission system based on access of unidirectional serial port channel |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131016 Termination date: 20150513 |
|
EXPY | Termination of patent right or utility model |