CN101150413A - A multi-frame cascading system and method for ATCA knife server - Google Patents
A multi-frame cascading system and method for ATCA knife server Download PDFInfo
- Publication number
- CN101150413A CN101150413A CNA2007101766412A CN200710176641A CN101150413A CN 101150413 A CN101150413 A CN 101150413A CN A2007101766412 A CNA2007101766412 A CN A2007101766412A CN 200710176641 A CN200710176641 A CN 200710176641A CN 101150413 A CN101150413 A CN 101150413A
- Authority
- CN
- China
- Prior art keywords
- blade server
- load balancing
- switch
- atca
- atca blade
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- JEOQACOXAOEPLX-WCCKRBBISA-N (2s)-2-amino-5-(diaminomethylideneamino)pentanoic acid;1,3-thiazolidine-4-carboxylic acid Chemical compound OC(=O)C1CSCN1.OC(=O)[C@@H](N)CCCN=C(N)N JEOQACOXAOEPLX-WCCKRBBISA-N 0.000 title claims abstract 42
- 238000012545 processing Methods 0.000 claims description 22
- 230000036541 health Effects 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims 1
- 238000004220 aggregation Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/24—Multipath
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Hardware Redundancy (AREA)
Abstract
This invention discloses a multi-frame cascade system for ATCA kife server, in which, each frame includes several exchange node plates and kife node plates, and the cascade system includes: a frame of host of the ATCA kife server, a frame of a sub-set and an exchanger, and the exchange node plates of the host frame and the sub-set frame are interconnected with the exchanger to realize multi-frame cascade of the ATCA kife server. This invention also provides a method for the cascade.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a multi-machine-frame cascading system and a method for an ATCA (advanced telecommunications Computing Architecture) blade server.
Background
The ATCA blade server is a frame type high-performance blade server constructed by adopting an ATCA (Advanced telecommunications computing architecture) specification. The blade server or blade server is a low-cost server platform with High Availability and High Density (HAHD), which is specially designed for special application industries and High-Density computer environments. The main structure of the blade server is a large main machine frame, and a plurality of plug-in cards in the shape of blades can be inserted inside the main machine frame, so that the plug-in cards are also called the blades for short, wherein each blade is actually a system motherboard, and is similar to an independent server, and an own operating system can be started through a local hard disk. Each blade can run its own system, serving a designated group of different users, without any association between them. The ATCA is a standard architecture system specification for next-generation carrier-grade communication equipment which is formulated and issued by a PICMG (Peripheral Component interconnect industrial Computer Manufacturers Group), evolved from a former picmg2.x and comprises a core specification PICMG3.0 and a series of auxiliary specifications picmg3.x, wherein the core specification defines parts such as mechanical design, system management, power distribution, heat dissipation, backplane interconnection and the like in the ATCA series specification; the auxiliary specification content defines the transport means of the interconnect in the core specification. The ATCA specification represents a fully open, modular industry standard, thus allowing telecommunication equipment manufacturers to employ interoperable, mature, commercial software and hardware components from third party vendors.
An ATCA blade server chassis typically includes multiple switch node boards and multiple high performance blade node boards. The switch node board supports two-layer switching, and the blade node board is used for realizing high-density calculation. Which may provide up to 12 blade node boards in an existing ATCA blade server chassis. When the computing power of the 12 blade node boards cannot meet the specific application requirements, a plurality of ATCA blade server machine frames need to be cascaded to form an ATCA blade server multi-machine frame cascading system.
At present, no relevant ATCA blade server multi-chassis cascading system construction scheme exists.
In summary, the technical solution of the conventional ATCA blade server multi-chassis cascading system obviously has inconvenience and defects in practical use, so it is necessary to improve the system.
Disclosure of Invention
In view of the above-mentioned drawbacks, a first object of the present invention is to provide a multi-shelf cascading system for an ATCA blade server, which can implement cascading of multiple ATCA blade server shelves.
The second purpose of the invention is to provide a method for cascading multiple frames of an ATCA blade server, which can realize the cascading of multiple ATCA blade server frames.
In order to achieve the first object, the present invention provides an ATCA blade server multi-chassis cascading system, where each ATCA blade server chassis includes several switching node boards and blade node boards, and the cascading system includes: the ATCA blade server comprises an ATCA blade server main frame, an ATCA blade server secondary frame and a switch;
and the switch node boards of the main machine frame and the sub machine frame of the ATCA blade server are interconnected with the switch to realize multi-machine-frame cascading of the ATCA blade server.
According to the cascading system, the ATCA blade server host shelf further comprises:
and the load balancing node board is interconnected with the switch and is used for shielding, scheduling and monitoring the blade node boards of the ATCA blade server main machine frame and the ATCA blade server secondary machine frame.
According to the cascading system, the load balancing node board provides a unified medium access control address and a virtual internet protocol address, each blade node board of the ATCA blade server main machine frame and the ATCA blade server sub machine frame has a private internet protocol address, and the internet protocol address is only visible for the load balancing node board to realize that the load balancing node board shields the blade node boards of the ATCA blade server main machine frame and the ATCA blade server sub machine frame.
According to the cascade system, the load balancing node board comprises:
the first load balancing node board is a main load balancing node board and is used for carrying out load balancing processing on the blade node boards of the main machine frame and the sub machine frame of the ATCA blade server when the ATCA blade server works normally and switching the blade node boards to the second load balancing node board when the ATCA blade server works abnormally;
and the second load balancing node board is a standby load balancing node board and is used for carrying out load balancing processing on the blade node boards of the ATCA blade server main machine frame and the ATCA blade server secondary machine frame after the first load balancing node board is switched.
According to the cascade system, the switch is a three-layer switch.
According to the cascade system, the switch further comprises: a first switch and a second switch;
the first switch and the second switch are in port convergence connection through a plurality of Ethernet interfaces;
the first switch and the second switch adopt a virtual routing redundancy protocol to realize the redundancy of an internet protocol layer; and/or the first switch and the second switch work in a main standby mode;
the first switch is connected with the first load balancing node boards one by one, and the second switch is connected with the second load balancing node boards one by one; and/or
The first switch is connected with the second load balancing node boards one by one, and the second switch is connected with the first load balancing node boards one by one.
According to the cascading system, the switching node boards of the ATCA blade server host frame and the ATCA blade server sub frame further include: a first switching node board and a second switching node board;
the first switching node board and the second switching node board work in a dual-active mode.
According to the cascade system, the first switching node board and the second switching node board are in cross interconnection with the first switch and the second switch, and the physical connection mode is port convergence connection through a plurality of Ethernet interfaces.
According to the cascade system, the first load balancing node board is switched to the second load balancing node board through a hardware main/standby switching signal when the first load balancing node board works abnormally, and meanwhile, load balancing data synchronization is carried out with the second load balancing node board through an Ethernet interface between the first load balancing node board and the second load balancing node board; and after the first load balancing node board is completely switched, the second load balancing node board sends an ARP (Address resolution protocol) notice to the outside, and performs load balancing processing on the blade node boards of the ATCA blade server main machine frame and the ATCA blade server sub machine frame.
According to the cascade system, the ethernet interface between the first load balancing node board and the second load balancing node board is a heartbeat ethernet interface.
In order to achieve the second object, the present invention provides an ATCA blade server multi-chassis cascading method, where each ATCA blade server chassis includes several switching node boards and blade node boards, and the method includes the following steps:
A. respectively setting the ATCA blade server multi-machine frame as an ATCA blade server main machine frame and an ATCA blade server sub-machine frame;
B. and interconnecting the exchange node boards of the ATCA blade server main frame and the ATCA blade server sub-frame with the exchanger.
According to the cascade method, the step a further comprises:
a1, respectively setting the ATCA blade server multi-chassis as an ATCA blade server main chassis and an ATCA blade server sub chassis;
and A2, inserting a load balancing node board on the ATCA blade server main frame to perform load balancing processing on the blade node boards of the ATCA blade server main frame and the ATCA blade server sub frame.
In the ATCA blade server multi-chassis cascading system, the ATCA blade server multi-chassis is set as the main machine chassis of the ATCA blade server and the secondary machine chassis of the ATCA blade server, and the mode that the switching node boards of the main machine chassis of the ATCA blade server and the secondary machine chassis of the ATCA blade server are in cross interconnection with the first layer switch and the second layer switch is adopted, so that the cascading of a plurality of ATCA blade server machine chassis is realized. In addition, the invention sets the first and the second load balancing node boards on the main frame of the ATCA blade server, the first and the second load balancing node boards and the first and the second three-layer switches are in the main and standby modes, the exchange node boards of the main frame of the ATCA blade server and the sub-frame of the ATCA blade server are in the double-main mode, each blade in the frame and the exchange node board of the frame are in cross connection through GE, and the double-network card is adopted to bind the same IP, thereby ensuring the high reliability and the safety of the multi-frame cascading system of the ATCA blade server.
Drawings
FIG. 1 is a schematic diagram of a multi-chassis cascading system of an ATCA blade server provided by the present invention;
FIG. 2 is a schematic diagram of a multi-chassis cascading system of a part of ATCA blade servers provided by the present invention;
FIG. 3 is a flow chart of a method for multi-chassis cascading of ATCA blade servers provided by the present invention;
figure 4 is a flowchart of the ATCA blade server multi-chassis cascading method provided by the preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The basic idea of the invention is: a three-layer switch and a load balancing node board are introduced into the ATCA blade server multi-chassis cascading system, so that the cascading of a plurality of ATCA blade server machine chassis is realized, and the high reliability and safety of the ATCA blade server multi-chassis cascading system are ensured.
As shown in fig. 1, the ATCA blade server multi-chassis cascading system 100 provided by the present invention includes a switch 101, an ATCA blade server main chassis 102, and an ATCA blade server sub-chassis 103. Wherein,
and the switch 101 is in cross interconnection with the ATCA blade server host frame 102 and the ATCA blade server sub-frame 103. Preferably, the switch 101 is a three-tier switch. The three-layer switch is a switch having a partial router function, and is superior to the two-layer and three-layer switches in performance, and can implement routing and forwarding of an IP packet once and many times by using an Application Specific Integrated Circuit (ASIC). As long as the function of a three-layer switch can be realized, it is within the scope of the present invention that the switch described in the present invention is a switch with more than three layers, i.e., a four-layer switch, a seven-layer switch, etc.
To improve the reliability of the multi-chassis cascade system 100 and eliminate single point failures, the switch 101 further includes a first switch 1011 and at least a second switch 1012, and the first switch 1011 and the second switch 1012 operate in the active/standby mode. The first switch 1011 is a main switch, and communicates with the ATCA blade server host machine frame 102 and the ATCA blade server sub machine frame 103 during normal operation, and switches to a standby switch, i.e., the second switch 1012, during abnormal operation. When a plurality of second switches 1012 exist, priorities can be set on the plurality of second switches 1012, and the first switch 1011 switches to an appropriate second switch 1012 according to the high-low selection of the priorities when the operation is abnormal.
The first switch 1011 and the second switch 1012 implement IP (Internet Protocol) layer redundancy by using VRRP (Virtual router redundancy Protocol), and the two-layer switch is a dual active mode. The first switch 1011 and the second switch 1012 are connected via a plurality of Ethernet interfaces to Trunk (port convergence) to increase the capacity of the physical link, so that the physical link can meet the requirement of bandwidth increase, and interface redundancy is realized, thereby improving the reliability of the system 100. The Ethernet interface is classified into a ten-mega Ethernet interface, a one-hundred-mega Ethernet interface, and a GE (Giga Ethernet) interface. In view of data transmission rate, the first switch 1011 and the second switch 1012 of the present invention are preferably connected via Trunk interfaces through a plurality of GE interfaces. It is within the scope of the present invention to implement Trunk interface connections via other ethernet interfaces. For convenience, the ethernet interface referred to below is preferably a GE interface.
An ATCA blade server chassis 102, completely independent and structurally different from the ATCA blade server chassis 103, the ATCA blade server chassis 103 including a switch node board 105 and a blade node board 106 but not including a load balancing node board 104, while the ATCA blade server chassis 102 includes a switch node board 105, a blade node board 106, and a load balancing node board 104. The load balancing node board 104 is interconnected with the switch 101 through a plurality of GE interfaces, and is configured to perform load balancing processing on the blade node boards 106 of the ATCA blade server main subrack 102 and the ATCA blade server subrack 103. The load balancing process includes: the blade node boards 106 of the ATCA blade server host shelf 102 and the ATCA blade server subshelf 103 are shielded, scheduled, and health monitored.
The load balancing node board 104 performs shielding, scheduling and health detection on each blade node board 106, so that the computing capability of the blade cluster can be fully exerted, and the management and expansion of the blade node boards 106 are facilitated.
Specifically, the load balancing node board 104 provides a uniform MAC (Media Access Control) address and a Virtual IP (Virtual Internet Protocol) address to the switch and a network other than the switch, and each of the ATCA blade server main chassis 102 and the ATCA blade server sub chassis 103 has a private IP (Internet Protocol) address which is visible only to the load balancing node board 104. Therefore, the load balancing node board 104 can shield the blade node boards 106 of the ATCA blade server main subrack 102 and the ATCA blade server secondary subrack 103, and high security of the system 100 can be ensured. Further, the present invention may integrate firewall functionality on the load balancing node board 104, thereby improving the security of the entire server system 100.
The load balancing node board 104 may periodically perform health monitoring on each blade node board 106 of the ATCA blade server main frame 102 and the ATCA blade server sub-frame 103, and when a faulty blade node board is detected, the faulty blade node board is removed from the frame. The load balancing node board 104 also supports session maintenance functions so that consistency of processing of associated sessions by each blade node board 106 can be ensured.
To improve the reliability of the multi-chassis cascading system 100 and eliminate single point failures, the load balancing node board 104 further includes: a first load balancing node board 1041 and a second load balancing node board 1042. Wherein,
the first load balancing node board 1041 is a master load balancing node board, and is configured to perform load balancing processing on the blade node boards 106 of the ATCA blade server master frame 102 and the ATCA blade server slave frame 103 when the first load balancing node board works normally, and switch to the second load balancing node board 1042 when the first load balancing node board works abnormally.
The second load balancing node board 1042 is a standby load balancing node board, and is configured to perform load balancing processing on the blade node boards 106 of the ATCA blade server host frame 102 and the ATCA blade server sub-frame 103 after the first load balancing node board 1041 is switched.
In the present invention, a heartbeat ethernet interface and a hardware host/standby line are provided between the first load balancing node board 1041 and the second load balancing node board 1042 to implement fast switching and data synchronization of the first load balancing node board 1041 and the second load balancing node board 1042, as shown in fig. 2. The term "heartbeat" is a communication statement, and is used herein to determine whether an ethernet interface is available between the first load balancing node board 1041 and the second load balancing node board 1042. The heartbeat ethernet interface in the present invention is preferably a heartbeat GE interface. The hardware main/standby line is used for providing a hardware main/standby switching signal. When the first load balancing node board 1041 is abnormal in operation, the first load balancing node board 1041 is switched to the second load balancing node board 1042 through the hardware main/standby switching signal, and meanwhile, load balancing data synchronization is performed with the second load balancing node board 1042 through the heartbeat GE interface; after the first load balancing node board 1041 is completely switched, the second load balancing node board 1042 sends an ARP (Address Resolution Protocol) announcement to the outside, the switch 101 updates its own switching table after receiving the announcement, and forwards the system data to the second load balancing node board 1042, and then the second load balancing node board 1042 performs load balancing processing on the blade node boards of the ATCA blade server main frame and the ATCA blade server sub frame.
As shown in fig. 1 and fig. 2, the first load balancing node boards 1041 are connected to the first switches 1011 one by one, and the second load balancing node boards 1042 are connected to the second switches 1012 one by one, and their physical connection mode is a Trunk interface connection through multiple GE interfaces. The invention can also connect the first switch 1011 and the second load balancing node 1042 one by one, and connect the second switch 1012 and the first load balancing node 1041 one by one, and the physical connection mode is to perform Trunk interface connection through multiple GE interfaces, which is also within the protection scope of the invention.
There is a switch node board 105 for both the ATCA blade server chassis 102 and the ATCA blade server sub chassis 103. To improve the reliability of the multi-chassis cascade system 100 and eliminate single point failures, the switching node boards 105 include a first switching node board 1051 and a second switching node board 1052, and the first switching node board 1051 and the second switching node board 1052 operate in dual active mode. In the present invention, the first switching node board 1051 and the second switching node board 1052 support two-layer ethernet switching and STP (Spanning Tree Protocol), so as to eliminate ethernet loops caused by cross connection with the switch 101 and realize redundancy. Preferably, the first switching node board 1051 and the second switching node board 1052 support two-layer ethernet switching and RSTP (Rapid Spanning tree protocol) to shorten the convergence time of the bridged network.
Referring to fig. 1 and 2, the first switching node board 1051 and the second switching node board 1052 are cross-connected with the first switch 1011 and the second switch 1012 in a physical connection manner of Trunk interface connection through a plurality of GE interfaces.
Similarly, both ATCA blade server chassis 102 and ATCA blade server sub-chassis 103 have multiple blade node boards 106, and the multiple blade node boards 106 are cross-interconnected with first switching node board 1051 and second switching node board 1052 via GE interfaces. In the invention, a main and standby double network card can be arranged on the plurality of blade node boards 106, the same IP is bound on the double network card, the network port driver adopts a physical link connection health detection and ARP health detection mode to detect the main network card, if the main network card is abnormal, the main and standby are switched, and the main network card is switched to the standby network card and then sends an ARP notification outwards.
Fig. 3 is a flowchart of an ATCA blade server multi-chassis cascading method provided by the present invention, the method is used for constructing the ATCA blade server multi-chassis cascading system 100 in fig. 1, where each ATCA blade server chassis includes a plurality of switch node boards 105 and a plurality of blade node boards 106, and the method specifically includes the following steps:
step S301, the ATCA blade server multi-machine frame is respectively set as an ATCA blade server main machine frame 102 and an ATCA blade server sub-machine frame 103.
The method further comprises the following steps: a load balancing node board 104 is inserted on the ATCA blade server host shelf 102 to load balance the blade node boards 106 of the ATCA blade server host shelf 102 and the ATCA blade server subshelf 103.
Step S302 is to interconnect the switch 101 with the switch node boards 105 of the ATCA blade server host shelf 102 and the ATCA blade server subshelf 103.
In this step, the switching node board 105 and the switch 101 are cross-connected, and the physical connection mode is through a Trunk interface of the GE interface.
Fig. 4 shows a method for multi-chassis cascading of an ATCA blade server according to a preferred embodiment of the present invention, where the method is used to construct an ATCA blade server multi-chassis cascading system 100 shown in fig. 1 and fig. 2, where each ATCA blade server chassis includes a switch node board 105 and N blade node boards 106, and the method specifically includes the following steps:
step S401, setting one ATCA blade server host machine frame 102 and other machine frames as ATCA blade server sub machine frames 103 in the ATCA blade server multi machine frame.
In step S402, a first load balancing node board 1041 and a second load balancing node board 1042 are inserted into the ATCA blade server host 102.
Step S403, setting 1 GE heartbeat interface and hardware main/standby switching line between the first load balancing node board 1041 and the second load balancing node board 1042.
Step S404, cross-connecting the switch node board 105 and the N blade node boards 106 on the ATCA blade server subrack 102 and the ATCA blade server subrack 103, where the physical connection mode is Trunk interface connection through the GE interface.
Step S405, setting a main and standby double network card on the switching node board 105 of the ATCA blade server machine frame 102 and the ATCA blade server machine frame 103, wherein the double network card binds the same IP.
The network port driver adopts physical link connection health detection and ARP health detection to detect the main network card, if the main network card is abnormal, the main network card is switched over, and the ARP notice is sent out after the main network card is switched over to the standby network card.
Step S406, respectively setting the first switching node board 1051 and the second switching node board 1052 on the switching node boards 105 of the ATCA blade server subrack 102 and the ATCA blade server subrack 103, and setting the operating modes of the first switching node board 1051 and the second switching node board 1052 to the dual active mode.
The first switching node board 1051 and the second switching node board 1052 each support a layer two ethernet switching and RSTP protocol.
Step S407, a first switch 1011 and a second switch 1012 are set, and cross-connected with a first switching node board 1051 and a second switching node board 1052, where the physical connection mode is Trunk interface connection through a GE interface.
Step S408 sets the first switch 1011 and the second switch 1012 to operate in the active/standby mode, and sets the first switch 1011 as the active switch and the second switch 1012 as the standby switch.
The VRRP protocol is adopted between the first switch 1011 and the second switch 1012 to realize IP layer redundancy and the two-layer switching works in a double-master mode.
Step S409, setting the first load balancing node board 1041 and the second load balancing node board 1042 to operate in a master-standby mode, and setting the first load balancing node board 1041 as a master load balancing node board and the second load balancing node board 1042 as a standby load balancing node board.
The load balancing node board 104 provides uniform MAC addresses and VIP addresses to the outside, the active load balancing node board monitors the connection status between the node board and the switch 103 through a protocol and a physical link detection mechanism, if the connection status is abnormal, the active and standby signals are switched through hardware, meanwhile, data synchronization is performed through the heartbeat GE interface, and after the switching is completed, the new active load balancing node sends out an ARP announcement.
In step S410, the first load balancing node board 1041 and the first switch 1011 are connected, and the second load balancing node board 1042 and the second switch 1012 are connected.
The physical connection mode between the first load balancing node board 1041 and the first switch 1011 and the physical connection mode between the second load balancing node board 1042 and the second switch 1012 are Trunk interface connection through a GE interface.
After the ATCA blade server multi-chassis cascading system 100 in fig. 1 and fig. 2 is constructed according to the cascading method shown in fig. 4, the data processing flow of the cascading system 100 is as follows:
when the first load balancing node board 1041 and the first switch 1011 operate normally, an external user accesses the ATCA blade server multi-chassis cascade system 100 through a VIP address provided by the load balancing node board, and a data stream is switched to the first load balancing node board 1041 in the main chassis through the first switch 1011. The first load balancing node board 1041 detects packet contents according to different configured load balancing functions, schedules a data packet at four layers or an application layer according to load conditions, health conditions and services that can be provided of each blade node board 106 of the current ATCA blade server subrack 102 and the ATCA blade server subrack 103, and uniformly forwards an external connection request to each blade node board 106 according to a certain scheduling policy, thereby achieving the purpose of load balancing. The first load balancing node board 1041 also periodically performs health detection on each blade node board 106, and removes a failed blade from the blade group.
When the first load balancing node board 1041 works abnormally, the first load balancing node board 1041 is switched to the second load balancing node board 1042, then an external user accesses the ATCA blade server multi-chassis cascade system 100 through a VIP address provided by the load balancing board, a data stream passes through the first switch 1011, passes through the third switch 1012, and passes through the second switch 1012 on the third switch 1012 to the second load balancing node board 1042 in the main chassis after the second switch is performed on the 1012. The specific data processing process is similar to the data processing process of the first load balancing node board 1041 and the first switch 1011 when the first load balancing node board 1041 and the first switch 1011 work normally, and is limited in space and is not expanded here, which may be specifically referred to as the data processing process of the first load balancing node board 1041 and the first switch 1011 when the first load balancing node board 1041 and the first switch 1011 work normally.
When the first switch 1011 works abnormally, the first switch 1011 is switched to the second switch 1012, then an external user accesses the ATCA blade server multi-chassis cascade system 100 through a VIP address, a data stream passes through the first switch 1011, passes through the second switch 1012, and is switched to the first switch 1011 after the third layer switching is performed on the second switch 1012, and the data stream is switched to the first load balancing node board 1041 in the main chassis through the second layer switching by the 1011. The specific data processing process is the same as the data processing process of the first load balancing node board 1041 and the first switch 1011 when the first load balancing node board 1041 and the first switch 1011 work normally, and is limited in space and is not expanded here, which may be specifically referred to as the data processing process of the first load balancing node board 1041 and the first switch 1011 when the first load balancing node board 1041 and the first switch 1011 work normally.
In summary, in the ATCA blade server multi-chassis cascade system of the present invention, the ATCA blade server multi-chassis is set as the ATCA blade server main chassis and the ATCA blade server sub chassis, and the ATCA blade server multi-chassis cascade is realized by interconnecting the first and second three-layer switches and the switching node boards of the ATCA blade server main chassis and the ATCA blade server sub chassis. In addition, the first and second load balancing node boards are arranged on the main frame of the ATCA blade server, the first and second load balancing node boards and the first and second three-layer switches are in a main standby mode, the switching node boards of the main frame of the ATCA blade server and the sub-frame of the ATCA blade server are in a double-main mode, blades in the frame and the switching node boards of the frame are in cross connection through GE, and the same IP is bound by adopting double-network cards, so that the flattening of the multi-frame cascading system of the ATCA blade server is realized, and the high reliability and the safety of the multi-frame cascading system of the ATCA blade server are ensured.
The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof, and it should be understood that various changes and modifications can be effected therein by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (12)
1. A kind of ATCA blade server multi-chassis cascade system, every ATCA blade server chassis includes several exchange node boards and blade node boards, characterized by that, the cascade system includes: the ATCA blade server comprises an ATCA blade server main frame, an ATCA blade server secondary frame and a switch;
and the switch node boards of the main machine frame and the sub machine frame of the ATCA blade server are interconnected with the switch to realize multi-machine-frame cascading of the ATCA blade server.
2. The cascading system of claim 1, wherein the ATCA blade server host shelf further comprises:
and the load balancing node board is interconnected with the switch and used for carrying out load balancing processing on the blade node boards of the ATCA blade server main frame and the ATCA blade server secondary frame, wherein the load balancing processing comprises shielding, scheduling and health monitoring on the blade node boards of the ATCA blade server main frame and the ATCA blade server secondary frame.
3. The cascading system of claim 2, wherein the load-balancing node board provides a unified media access control address and a virtual internet protocol address, and wherein each blade node board of the ATCA blade server host shelf and the ATCA blade server subshelf has a private internet protocol address that is visible only to the load-balancing node board to enable the load-balancing node board to mask the blade node boards of the ATCA blade server host shelf and the ATCA blade server subshelf.
4. The cascade system of claim 3, wherein the load balancing node board comprises:
the first load balancing node board is a main load balancing node board and is used for carrying out load balancing processing on the blade node boards of the main machine frame and the sub machine frame of the ATCA blade server when the ATCA blade server works normally and switching the blade node boards to the second load balancing node board when the ATCA blade server works abnormally;
and the second load balancing node board is a standby load balancing node board and is used for carrying out load balancing processing on the blade node boards of the ATCA blade server main machine frame and the ATCA blade server secondary machine frame after the first load balancing node board is switched.
5. The cascade system of claim 4, wherein the switch is a three-tier switch.
6. The cascade system of claim 5, wherein the switch further comprises: a first switch and a second switch;
the first switch and the second switch are in port convergence connection through a plurality of Ethernet interfaces;
the first switch and the second switch adopt a virtual routing redundancy protocol to realize the redundancy of an internet protocol layer; and/or the first switch and the second switch work in a main standby mode;
the first switch is connected with the first load balancing node boards one by one, and the second switch is connected with the second load balancing node boards one by one; and/or
The first switch is connected with the second load balancing node boards one by one, and the second switch is connected with the first load balancing node boards one by one.
7. The cascading system of claim 6, wherein the switching node boards of the ATCA blade server host shelf and the ATCA blade server sub-shelf further comprise: a first switching node board and a second switching node board;
the first switching node board and the second switching node board work in a dual-active mode.
8. The cascading system of claim 7, wherein the first and second switch node boards are cross-connected to the first and second switches in a physical connection manner that is port aggregation connection through a plurality of ethernet interfaces.
9. The cascade system according to claim 4, wherein the first load balancing node board switches to the second load balancing node board through the hardware main/standby switching signal when the operation is abnormal, and performs load balancing data synchronization with the second load balancing node board through an ethernet interface between the first load balancing node board and the second load balancing node board; and after the first load balancing node board is completely switched, the second load balancing node board sends an ARP (Address resolution protocol) notice to the outside, and performs load balancing processing on the blade node boards of the ATCA blade server main machine frame and the ATCA blade server sub machine frame.
10. The cascading system of claim 9, wherein the ethernet interface between the first load-balancing node board and the second load-balancing node board is a heartbeat ethernet interface.
11. A method of constructing an ATCA blade server multi-chassis cascade system according to any of claims 1-10, each ATCA blade server chassis comprising a plurality of switch node boards and blade node boards, comprising the steps of:
A. respectively setting the ATCA blade server multi-machine frame as an ATCA blade server main machine frame and an ATCA blade server sub-machine frame;
B. and interconnecting the exchange node boards of the ATCA blade server main frame and the ATCA blade server sub-frame with the exchanger.
12. The cascade method of claim 11, wherein the step a further comprises:
a1, respectively setting the ATCA blade server multi-chassis as an ATCA blade server main chassis and an ATCA blade server sub chassis;
and A2, inserting a load balancing node board on the ATCA blade server main frame to perform load balancing processing on the blade node boards of the ATCA blade server main frame and the ATCA blade server sub frame.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2007101766412A CN101150413B (en) | 2007-10-31 | 2007-10-31 | A multi-frame cascading system and method for ATCA knife server |
PCT/CN2007/003694 WO2009055982A1 (en) | 2007-10-31 | 2007-12-19 | An atca blade server multi-frame concatenation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2007101766412A CN101150413B (en) | 2007-10-31 | 2007-10-31 | A multi-frame cascading system and method for ATCA knife server |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101150413A true CN101150413A (en) | 2008-03-26 |
CN101150413B CN101150413B (en) | 2010-06-02 |
Family
ID=39250766
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2007101766412A Active CN101150413B (en) | 2007-10-31 | 2007-10-31 | A multi-frame cascading system and method for ATCA knife server |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN101150413B (en) |
WO (1) | WO2009055982A1 (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010072175A1 (en) * | 2008-12-25 | 2010-07-01 | 华为技术有限公司 | Multi-frame interconnection system and frame connection method |
CN101853185A (en) * | 2009-03-30 | 2010-10-06 | 华为技术有限公司 | Blade server and service dispatching method thereof |
CN101895444A (en) * | 2010-07-28 | 2010-11-24 | 南京信息工程大学 | Dual system of ATCA blade server, connection method and test method |
CN101345631B (en) * | 2008-08-14 | 2011-03-02 | 中兴通讯股份有限公司 | Management method and system for blade server system exchange plane |
CN101557294B (en) * | 2009-05-08 | 2011-11-30 | 中兴通讯股份有限公司 | Machine frame cascading networking system of ATCA blade server and method thereof |
WO2011150746A1 (en) * | 2010-06-01 | 2011-12-08 | 中兴通讯股份有限公司 | Cascade system and method for realizing automatic circuit bridge connection |
CN101610283B (en) * | 2009-07-17 | 2012-04-25 | 浪潮电子信息产业股份有限公司 | Method for dynamically balancing received load under network card binding state |
CN103023973A (en) * | 2012-11-19 | 2013-04-03 | 浪潮电子信息产业股份有限公司 | Cluster server designing method based on CPCI (Compact Peripheral Component Interconnect) structure |
CN103106126A (en) * | 2013-01-16 | 2013-05-15 | 浪潮电子信息产业股份有限公司 | High-availability computer system based on virtualization |
CN104539558A (en) * | 2014-12-31 | 2015-04-22 | 林坚 | Capacity-expansible IP telephone exchange blade mechanism frame and automatic capacity expansion method |
CN104639367A (en) * | 2015-01-04 | 2015-05-20 | 中国联合网络通信集团有限公司 | Method and system for implementing switching of master server and standby server |
CN104679714A (en) * | 2015-03-10 | 2015-06-03 | 江苏微锐超算科技有限公司 | Supercomputer cluster based on ATCA (advanced telecom computing architecture) |
CN107968724A (en) * | 2017-11-19 | 2018-04-27 | 中国船舶重工集团公司第七六研究所 | Gigabit, 100,000,000 dual redundant Ethernet cards spare part system and method each other |
CN109788012A (en) * | 2017-11-14 | 2019-05-21 | 阿里巴巴集团控股有限公司 | A kind of health examination method and device, health examination control method and controller |
TWI687802B (en) * | 2018-01-30 | 2020-03-11 | 廣達電腦股份有限公司 | Method for ensuring normal operation of a power supply unit in an electronic device |
CN112087531A (en) * | 2020-08-14 | 2020-12-15 | 陕西千山航空电子有限责任公司 | Method for on-line configuring network address |
CN112486893A (en) * | 2020-12-08 | 2021-03-12 | 中国电子科技集团公司第三十研究所 | Computing server system architecture |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7245632B2 (en) * | 2001-08-10 | 2007-07-17 | Sun Microsystems, Inc. | External storage for modular computer systems |
US20040081104A1 (en) * | 2002-10-29 | 2004-04-29 | Weimin Pan | Method and system for network switch configuration |
US7480720B2 (en) * | 2003-06-25 | 2009-01-20 | International Business Machines Corporation | Method and system for load balancing switch modules in a server system and a computer system utilizing the same |
US20060190484A1 (en) * | 2005-02-18 | 2006-08-24 | Cromer Daryl C | System and method for client reassignment in blade server |
US20060203715A1 (en) * | 2005-03-14 | 2006-09-14 | International Business Machines Corporation | Method for redirection of virtual LAN network traffic |
CN2852260Y (en) * | 2005-12-01 | 2006-12-27 | 华为技术有限公司 | Server |
-
2007
- 2007-10-31 CN CN2007101766412A patent/CN101150413B/en active Active
- 2007-12-19 WO PCT/CN2007/003694 patent/WO2009055982A1/en active Application Filing
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101345631B (en) * | 2008-08-14 | 2011-03-02 | 中兴通讯股份有限公司 | Management method and system for blade server system exchange plane |
WO2010072175A1 (en) * | 2008-12-25 | 2010-07-01 | 华为技术有限公司 | Multi-frame interconnection system and frame connection method |
US8619764B2 (en) | 2008-12-25 | 2013-12-31 | Huawei Technologies Co., Ltd. | System for multi-chassis interconnection and method for chassis connection |
US8527565B2 (en) | 2009-03-30 | 2013-09-03 | Huawei Technologies Co., Ltd. | Selecting and reassigning a blade for a logical partition for service scheduling of a blade server |
CN101853185B (en) * | 2009-03-30 | 2015-07-08 | 华为技术有限公司 | Blade server and service dispatching method thereof |
CN101853185A (en) * | 2009-03-30 | 2010-10-06 | 华为技术有限公司 | Blade server and service dispatching method thereof |
CN101557294B (en) * | 2009-05-08 | 2011-11-30 | 中兴通讯股份有限公司 | Machine frame cascading networking system of ATCA blade server and method thereof |
CN101610283B (en) * | 2009-07-17 | 2012-04-25 | 浪潮电子信息产业股份有限公司 | Method for dynamically balancing received load under network card binding state |
WO2011150746A1 (en) * | 2010-06-01 | 2011-12-08 | 中兴通讯股份有限公司 | Cascade system and method for realizing automatic circuit bridge connection |
CN101895444A (en) * | 2010-07-28 | 2010-11-24 | 南京信息工程大学 | Dual system of ATCA blade server, connection method and test method |
CN103023973A (en) * | 2012-11-19 | 2013-04-03 | 浪潮电子信息产业股份有限公司 | Cluster server designing method based on CPCI (Compact Peripheral Component Interconnect) structure |
CN103106126A (en) * | 2013-01-16 | 2013-05-15 | 浪潮电子信息产业股份有限公司 | High-availability computer system based on virtualization |
CN104539558A (en) * | 2014-12-31 | 2015-04-22 | 林坚 | Capacity-expansible IP telephone exchange blade mechanism frame and automatic capacity expansion method |
CN104539558B (en) * | 2014-12-31 | 2018-09-25 | 林坚 | Extendible capacity IP phone interchanger rolling reamer machine system and automatic expansion method |
CN104639367A (en) * | 2015-01-04 | 2015-05-20 | 中国联合网络通信集团有限公司 | Method and system for implementing switching of master server and standby server |
CN104639367B (en) * | 2015-01-04 | 2018-04-03 | 中国联合网络通信集团有限公司 | A kind of method and system for realizing active/standby server switching |
CN104679714A (en) * | 2015-03-10 | 2015-06-03 | 江苏微锐超算科技有限公司 | Supercomputer cluster based on ATCA (advanced telecom computing architecture) |
CN109788012A (en) * | 2017-11-14 | 2019-05-21 | 阿里巴巴集团控股有限公司 | A kind of health examination method and device, health examination control method and controller |
CN109788012B (en) * | 2017-11-14 | 2022-04-01 | 阿里巴巴集团控股有限公司 | Health examination method and device, health examination control method and controller |
CN107968724A (en) * | 2017-11-19 | 2018-04-27 | 中国船舶重工集团公司第七六研究所 | Gigabit, 100,000,000 dual redundant Ethernet cards spare part system and method each other |
CN107968724B (en) * | 2017-11-19 | 2021-02-12 | 中国船舶重工集团公司第七一六研究所 | System and method for gigabit and hundred million double-redundancy Ethernet cards to be spare parts of each other |
TWI687802B (en) * | 2018-01-30 | 2020-03-11 | 廣達電腦股份有限公司 | Method for ensuring normal operation of a power supply unit in an electronic device |
CN112087531A (en) * | 2020-08-14 | 2020-12-15 | 陕西千山航空电子有限责任公司 | Method for on-line configuring network address |
CN112486893A (en) * | 2020-12-08 | 2021-03-12 | 中国电子科技集团公司第三十研究所 | Computing server system architecture |
Also Published As
Publication number | Publication date |
---|---|
WO2009055982A1 (en) | 2009-05-07 |
CN101150413B (en) | 2010-06-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101150413B (en) | A multi-frame cascading system and method for ATCA knife server | |
CN101588304B (en) | Implementation method of VRRP and device | |
US6253266B1 (en) | Apparatus and method for controlling information flow in a card cage having multiple backplanes | |
EP1721424B1 (en) | Interface bundles in virtual network devices | |
EP2617165B1 (en) | System and method for providing ethernet over infiniband virtual hub scalability in a middleware machine environment | |
CN102439903B (en) | Method, device and system for realizing disaster-tolerant backup | |
CN101383835B (en) | Method and device for implementing server safe isolation | |
WO2007092132A2 (en) | System and method for detecting and recovering from virtual switch link failures | |
CN1938990A (en) | Advanced switching fabric discovery protocol | |
US20110145630A1 (en) | Redundant, fault-tolerant management fabric for multipartition servers | |
CN102006189A (en) | Primary access server determination method and device for dual-machine redundancy backup | |
CN113839862B (en) | Method, system, terminal and storage medium for synchronizing ARP information between MCLAG neighbors | |
CN104468304B (en) | A kind of method of pond elementary state synchronizing information, pond Register and pond element | |
CN101360043B (en) | Communication apparatus reliably connected inside | |
CN100426743C (en) | Internal reliable interconnect communication device | |
WO2016106972A1 (en) | Data exchange method, multi-frame interconnection system, and frame device thereof | |
CN104753707A (en) | System maintenance method and network switching equipment | |
CN202940854U (en) | Multi-I/O network application system based on public computing service | |
CN101895444A (en) | Dual system of ATCA blade server, connection method and test method | |
EP2538612A1 (en) | Method and device for managing service | |
CN102291313B (en) | Method, device and equipment for finding virtual router redundancy protocol (VRRP) network topology | |
CN108206748B (en) | Core network backup method applied to TD-LTE (time division-Long term evolution) Internet of things service isolation environment | |
CN113805788B (en) | Distributed storage system and exception handling method and related device thereof | |
CN114124803A (en) | Device management method, device, electronic device and storage medium | |
CN109600270B (en) | Network equipment control method and network equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |