`
`(12) United States Patent
`US 6,978,346 B2
`(10) Patent N0.:
` Baek et al.
`(45) Date of Patent:
`Dec. 20, 2005
`
`(54) APPARATUS FOR REDUNDANT
`INTERCONNECTION BETWEEN MULTIPLE
`HOSTS AND RAID
`
`................ 714/4
`8/2003 Nguyen et al.
`6,609,213 B1 *
`............... 711/114
`6,820,171 B1 * 11/2004 Weber et al.
`OTHER PUBLICATIONS
`
`(75)
`
`Inventors: Sung-Hoon Baek, Taejon (KR);
`Joong-Bae Kim, Taejon (KR);
`Yong-Youn Kim, Taejon (KR)
`
`(73) Assignee: Electronics and Telecommunications
`Reseach Institute, (KR)
`
`( * ) Notice:
`
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 856 days.
`
`(21) Appl. N0.: 09/753,245
`
`(22) Filed:
`
`Dec. 29, 2000
`
`(65)
`
`Prior Publication Data
`
`US 2002/0035669 A1
`
`Mar. 21, 2002
`
`(30)
`
`Foreign Application Priority Data
`
`Sep. 19, 2000
`
`(KR)
`
`............................... 2000-54807
`
`Int. Cl.7 ......................... G06F 13/00; G06F 12/00
`(51)
`(52) US. Cl.
`....................................... 711/114; 709/250
`(58) Field of Search ................................ 709/201—203,
`709/217—219, 223—224, 239—240, 244, 250;
`711/114; 710/38; 370/360, 412
`
`(56)
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`g’ggi’gzg 2
`5,798,306 A
`5,812,754 A *
`6,192,485 B1 *
`
`1011;113:2112: Z: al‘
`13/133;
`8/1998 Dickinson, Jr.
`9/1998 Lui et al.
`....................... 714/6
`2/2001 Takita et al.
`................... 714/6
`
`Tellurite glass: a new candidate for fiber devices by J .S.
`Wang: Optical Materials 3 (1994) 187-203.
`Structure and optical properties of rare earth doped Zinc
`oxyhalide tellurite glasses by D.L. Sidebottom: Journl of
`Non-Crystalline Solids 222 (1997), pp. 282-289.
`Raman spectra snd thermal analysis of a new lead-tellurium-
`germanate glass system by Z. Pan: Journl of Non-Crystalline
`Solids 210 (1997), pp. 130-135.
`
`* cited by examiner
`
`Primary Examiner—Krisna Lim
`(74) Attorney, Agent, or Firm—Blakely Sokoloff Taylor &
`Zafman
`
`(57)
`
`ABSTRACT
`
`interconnection between
`The apparatus for a redundant
`multiple hosts and a redundant array of inexpensive disks
`(hereinafter, referred to as ‘RAID’), which is capable of
`supporting a fault tolerance of RAID controllers and simul-
`taneously heightening a performance, comprises a plurality
`of RAID controlling units for processing a requirement of
`numerous host computers connected with one another
`through the industrial standard communication network and
`for fault tolerance; a plurality of connecting units for con-
`necting the plurality of RAID controlling units to the numer-
`ous host computers; and a plural number of network inter-
`face controlling units
`respectively contained into the
`plurality of RAID controlling units, for exchanging infor-
`mation directly with an opposite network interface control-
`ling unit provided within an opposite RAID controlling unit
`and the numerous host computers, through the plurality of
`connecnng umts'
`
`9 Claims, 6 Drawing Sheets
`
`______ 119.5
`______132’______ .1.422 ______ 1.423,. ______ 1.494
`, ______ 139°
`I
`[HOST COMPUTERI
`IHDST COMPUTERl
`IrHOST COMPUTERIIHOST COMPUTERI IrHOST COMPUTERl
`HOST COMPUTERI
`1
`
`i ”11'. . r
`1
`7 vv .-
`411I
`412 II
`ml
`414 I
`410
`415
`i
`I
`V 7 ’W. - v
`:
`INTERFACE
`INTERFACE
`INTERFACE
`CONTROLLER
`CONTROLLER
`CONTROLLER
`
`I._..___..___
`
`l.____—_..—
`
`,._______-
`
`ll I
`
`Il
`|
`
`|_________
`
`,____.._..._-
`
`EH;
`l'nmdf'm'm‘fl
`
`l'
`
`_ ‘l
`
`NETWORK INTERFACE NETWORK INTERFACE
`CONTROLLER
`481
`RAID CONTROLLER
`L_________________________
`
`DHPN-1001 /Page 1 0f10
`
`
`
`US. Patent
`
`Dec. 20, 2005
`
`Sheet 1 0f 6
`
`US 6,978,346 B2
`
`F I G.
`
`1
`
`-—PRIOR ART--
`
`-1
`
`I I I I
`
`III I I
`
`.J
`
`101
`l— __________ ‘f— _ —
`HOST COMPUTER
`
`111
`NETWORK INTERFACE
`
`100
`I— __________ “L _ _ "I
`l
`I
`HOST COMPUTER
`l
`I
`I
`I
`|
`I
`I
`I
`I
`I
`I
`I
`I
`I
`
`110
`NETWORK INTERFACE
`CONTROLLER
`
` CONTROLLER
`
`NETWORK INTERFACE
`CONTROLLER
`
`NETWORK INTERFACE
`CONTROLLER
`
`DHPN-1001 / Page 2 of 10
`
`
`
`US. Patent
`
`Dec. 20, 2005
`
`Sheet 2 0f 6
`
`US 6,978,346 B2
`
`FIG. 2
`~~PRIOR ART--
`
`J 200
`
`—————————— —/—2—01—
`
`HOST COMPUTER
`
`
`
`NETWORK INTERFACE
`NETWORK INTERFACE
`CONTROLLER
`CONTROLLER
`
`HUB OR SWITCH
`
`
`
`
`
`
`
`
`
`
`F——___"———_'I
`ll’F'EH
`
`
`
`
`
`
`COMMUNI-
`COMMUNI-
`NETWORK
`
`CATION
`CATION
`INTERFACE
`INTERFACE
`
`
`
`CONTROLLER
`CONTROLLER
`CONTROLLER
`CONTROLLER
`
`
`
`
`
`
`220
`221
`RAID CONTROLLER
`
`
`DHPN-1001 / Page 3 0f10
`
`
`
`US. Patent
`
`Dec. 20, 2005
`
`Sheet 3 0f 6
`
`US 6,978,346 B2
`
`FIG. 3
`--PRIOR ART--
`
`NETWORK INTERFACE
`CONTROLLER
`
`NETWORK INTERFACE
`CONTROLLER
`
`CONTROLLER
`
`NEIWORK
`
`INTERFACE
`CONTROLLER
`
`INTERFACE
`
`DHPN-1001 / Page 4 0f10
`
`
`
`US. Patent
`
`Dec. 20, 2005
`
`Sheet 4 0f 6
`
`US 6,978,346 B2
`
`_.
`
`__
`
`«MII
`
`“E3828"BEE?
`
`_Iwo<mmw._.2_"___
`
`I...
`
`”SHNSH§\81:v0.7.5528EOIL@5068SOIL@228.$0er@228BEL..rIwSLIooBOIL.ESIIooEOILm8\31:
`
`
`
`
`
`
`
` mijEzOo
`mofimflz"__________Z_m;_Em:N;E.o;
`IoEsmIoIIIAhHM1MIrHuAlxlmllle‘Fm—E
`
`
`
`.1llllllllllllllllllllllllllllllllllllllllllllllllllJ5*£565289%4_.5$8589%on.
` _.___I___ImjoEzoo:ImjoEzoo:$30528$50528nmoEIEz.:moEIEz:moEIEz.
`
`
`
`
`“a“!Huh”Fig-m“
`
`EHI856IoIII
`
`DHPN-1001 / Page 5 0f10
`
`
`
`
`
`US. Patent
`
`Dec. 20, 2005
`
`Sheet 5 0f 6
`
`US 6,978,346 B2
`
`m.GE
`
`
`
`LFII
`”Bx:
`
`LFlu:
`Nomfi
`
`|_rll
`511
`
`_mmzmeOO50:__cwHDQEOOHwOI__mmCEEOO50:__mwham§oo.50:
`
`mom\
`
` _xmozfiwz“_rmwSmEOWHwOIL
`
`
`wOEmprI
`wo<mmmpz_
`
`mijmHzOo
`
`mijmHzOo
`
` —I
`Iotammom5:IXHE1IXEE
`
`
`||||r__||||
`
`EH
`
`__I
`
`VEOBEZ
`
`wo<umw._.z_
`
`__
`
`__
`
`__
`
`meEEz
`
`mo<uEsz_
`
`mijmkzOo
`:EjoEzoo
`
`________J
`
`vEQSEZ
`
`wO/EmmHZ
`
`cijmHzoo
`
`mo<mmm_._.z_
`
`mijEPzOO
`
`j..........I“?!
`
`
`.I'u\
`_|||||
`
`DHPN-1001 / Page 6 0f10
`
`
`
`IXHEL‘IXEE
`
`
`
`
`
`
`
`
`
`
`
`US. Patent
`
`2
`
`2B643,879,6SU
`
`fl-:11:1_D________
`
`
`
`
`|__Il.l.\ullllllL_||.!\u|||||||_MEN#8m8.N85m08mGE
`EjoEzoo_moEEE_.|_BEE?__moEEE__BEE?__wofimflz_onEE._xmoamz"“xmozfiz:x5252:xmogmz:nnE8228Bo:_E32200501__6:32850:__Esazoo50:__$512850:_E2828Go:__||.lllllll|__|||\IIIIIIIL_|l|\.IIIIII|:|||\:llllll
`EjoEzoo“EjoEzoo:$30528:EjoEzoo:EjoEzoo
`
`5%—EE-5%n0llllllllllllllllllllllllllllllll
`
`
`_|llllllllllllllllllllllllllllllllI.mEafigEaEa“6“a:m_e_h_S_M_m,—EE-
`
`___
`
`DHPN-1001 / Page 7 0f10
`
`
`
`
`US 6,978,346 B2
`
`1
`APPARATUS FOR REDUNDANT
`INTERCONNECTION BETWEEN MULTIPLE
`HOSTS AND RAID
`
`FIELD OF THE INVENTION
`
`The present invention relates to an apparatus for a redun-
`dant interconnection between multiple host computers and a
`redundant arrays of inexpensive disks (hereinafter, referred
`to as ‘RAID’); and, more particularly, to an apparatus for a
`redundant interconnection between multiple host computers
`and multiple controllers of the RAID, which is capable of
`supporting a fault tolerance of the RAID controllers and
`simultaneously heightening performance.
`
`PRIOR ART OF THE INVENTION
`
`ARAID is a storage system based on a large capacity and
`a high performance, by using much quantity of disks, and is
`a fault tolerant system in which the disks or controllers etc.
`have a redundant nature. In general,
`the RAID has two
`controllers, which are used like a method shown in FIG. 1
`or 2.
`
`FIG. 1 is an exemplary block diagram showing a general
`connection method between the host computers and the
`RAID having the conventional two controllers.
`As shown in the drawing, the RAID 130 includes two
`RAID controllers 140, 141 and each of RAID controllers
`140, 141 includes network interface controllers 150, 151.
`The network interface controllers 150, 151 of the RAID
`controllers 140, 141 are independently connected to network
`interface controllers 110, 111 of the host computers 100, 101
`through communication links 120, 121 such as a copper line
`and an optical fiber. That is, such system has twice the
`bandwidth and twice the performance. However, there is
`such a problem that a loss of data occurs when one out of
`two RAID controllers 140, 141 has a trouble, in other words,
`this system does not become the fault tolerant system.
`FIG. 2 is an exemplary block diagram of a general host
`interface system having a communication interface for an
`error recovery between the conventional two controllers.
`In order to provide fault tolerance not provided in FIG. 1,
`RAID 240 includes two RAID controllers 230, 231 and two
`RAID controllers 230, 231 and host computers 200, 201 are
`connected with each other through a hub or switch 210 in
`one network. The RAID controller 230 includes a pair of
`network interface controllers 220 and 221 and the RAID
`
`controller 231 includes a pair of network interface control-
`lers 222 and 223. Thus, even though one RAID controller
`230 or 231 has a trouble, all of the host computers 200, 201
`are connected to a RAID controller that does not have a
`
`trouble. That is, this RAID controller not having the trouble
`serves as a role of the controller that has the trouble. Also,
`since the RAID controllers 230, 231 should exchange infor-
`mation with each other by preparing in advance against
`some trouble, the RAID controllers 230, 231 are connected
`with each other through communication controllers 221,
`222. However, in this case only a half of performance for the
`bandwidth provided in FIG. 1 can be obtained.
`FIG. 3 is an exemplary block diagram showing a wiring
`method between a conventional RAID and the host com-
`puters.
`The construction shown in the drawing partially repre-
`sents a systematic connection between a RAID and host
`computers, which is extracted from contents disclosed in the
`US. Pat. No. 5,812,754. The RAID 340 includes two RAID
`controllers each of which has network interference control-
`
`10
`
`15
`
`20
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`lers 330, 331 and four ports 310, 311, 320 and 321. How-
`ever, this construction has no any difference from that of
`FIG. 2, in the structure of a communication network, and in
`case that one out of two host computers 300, 301 has rather
`a trouble, there is caused a problem that a network is broken.
`Thus, this construction is inferior to the construction of FIG.
`2.
`
`SUMMARY OF THE INVENTION
`
`invention to
`is an object of the present
`it
`Therefore,
`provide an apparatus for
`a
`redundant
`interconnection
`between multiple host computers and a RAID, which is
`capable of supporting a fault tolerance of a RAID controller
`and simultaneously heightening a performance.
`In accordance with the present invention, the apparatus
`for a redundant interconnection between multiple hosts and
`a RAID comprises a plurality of RAID controllers for
`processing requests of numerous host computers connected
`with one another through an industrial standard communi-
`cation network such as fibre channel and performing fault
`tolerant function; a plurality of connecting units for con-
`necting the plurality of RAID controllers to the numerous
`host computers; and a plural number of network interface
`controllers respectively contained into the plurality of RAID
`controllers,
`the network interface controllers being for
`exchanging information directly with each of opposite net-
`work interface controllers provided within the numerous
`host computers and within opposite RAID controllers,
`through the plurality of connecting units.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`The above and other objects and features of the instant
`invention will become apparent from the following descrip-
`tion of preferred embodiments taken in conjunction with the
`accompanying drawings, in which:
`FIG. 1 is an exemplary block diagram showing a general
`connection system between host computers and a RAID
`having conventional two controllers;
`FIG. 2 indicates an exemplary block diagram of a general
`host interface system having a communication interface for
`an error recovery between the conventional two controllers;
`FIG. 3 illustrates an exemplary block diagram of a wiring
`method between a conventional RAID and host computers;
`FIG. 4 is a block diagram showing one embodiment of a
`host
`interface system as an internal
`installment system
`between a RAID and host computers in accordance with the
`present invention;
`FIG. 5 depicts a block diagram providing one embodi-
`ment of a host interface system as an external installment
`system between a RAID and host computers in the present
`invention; and
`FIG. 6 is a block diagram showing one embodiment of a
`host interface system as a network switch between a RAID
`and host computers in the invention.
`
`PREFERRED EMBODIMENT OF THE
`INVENTION
`
`Hereinafter, preferred embodiments of the present inven-
`tion will be described in detail with reference to the accom-
`
`panying drawings.
`FIG. 4 is a block diagram showing one embodiment of a
`host matching system as an internal
`installment system
`between a RAID and host computers in accordance with the
`present invention.
`
`DHPN-1001 / Page 8 0f10
`
`
`
`US 6,978,346 B2
`
`3
`interface
`in the inventive host
`As shown in FIG. 4,
`system, a communication circuit is provided in order for an
`error recovery between two RAID controllers 460, 461, and
`the bandwidth between two groups as the host computers
`400 to 405 and two RAID controllers 460, 461 becomes
`twice the single connection bandwidth. Also, in the inven-
`tive host interface system, even though one RAID controller
`460 or 461 has an occurrence of a trouble, the bandwidth
`becomes twice the single connection bandwidth.
`That is to say, in a RAID 490, two RAID controllers 460,
`461 and hubs 440, 441 exist, and in each of the RAID
`controllers 460, 461, a pair of network interface controllers
`470, 471; 480, 481 are provided. Herewith, the hubs 440,
`441 are provided to connect a system connected to these
`hubs by one network and maintain the network even though
`one system has an occurrence of a trouble or a short of a line,
`and it can be as a hub or a switch. Hereinafter, they are
`named a “hub” altogether.
`Hub ports, 420 to 424, 430 to 434, shown in FIG. 4
`indicate an example for a simple internal structure of a fibre
`channel arbitrated loop hub, and this is based on an already
`well-known technique, thus there will be herein no more
`description therefore in the invention. The hub observes its
`corresponding communication network standard.
`A network, in which the RAID controllers, the hubs and
`the host computers are connected with one another, corre-
`sponds to the industrial standard communication network
`such as fibre channel, asynchronous transfer mode (ATM)
`and InfiniBand etc. and they are hereinafter named a ‘net-
`work’.
`Network interface controllers, 410 to 415, contained into
`the host computers, 400 to 405, and the network interface
`controllers 470, 471, 480, 481 of the RAID controllers 460,
`461 are connected with one another by two networks
`through two hubs 440, 441, and according to a sort of the
`networks, the network interface controller becomes a fibre
`channel controller, an ATM controller and an InfiniBand
`controller etc.
`
`representatively
`this time, a communication line,
`At
`shown as 450 in the drawing, for connecting the network
`interface controller to the hub is a copper line or an optical
`fibre, which is matched to a corresponding standard.
`Meanwhile, two network interface controllers 470, 471 of
`the first RAID controller 460 are respectively connected to
`two different hub ports 423, 432, and two network interface
`controllers 480, 481 of the second RAID controller 461 are
`respectively connected to two different hub ports 422, 433.
`The rest ports 420, 421, 424, 430, 431, 434 of the hubs 440,
`441 are connected to the host computers 400 to 405. Just,
`there is no distinction between the hub ports 420 to 424 of
`the first hub 440 at all. Also, there is no distinction between
`the hub ports 430 to 434 of the second hub 441 at all.
`The hub port connected to the host computer among the
`hub ports of the hub 440, namely, 420, 421, 424, is more
`than one, and there is no limitation to the maximum number.
`Further, What it is connected to the host computer among the
`hub ports of the second hub 441, namely, 430, 431, 434, is
`more than one, and there is no limitation to the maximum
`number. The hub ports 424, 434 and the host computers 400,
`405, which are shown as dot lines in FIG. 4, mean that there
`is no, or more than one hub port or host computer.
`Since, in such construction, two independent networks are
`constructed; it has twice the bandwidth of the single net-
`work, and a communication passage between two RAID
`controllers needed to perform the fault tolerant function of
`two RAID controllers 460, 461 is formed. Thus, information
`from the second network interface controller 471 of the first
`
`4
`RAID controller 460 is sent to the first network interface
`controller 481 of the second RAID controller 461. Also,
`information from the second network interface controller
`480 of the second RAID controller 461 is transmitted to the
`first network interface controller 470 of the first RAID
`controller 460. Further, information from the first network
`interface controller 481 of the second RAID controller 461
`is transmitted to the second network interface controller 471
`of the first RAID controller 460, and information from the
`first network interface controller 470 of the first RAID
`controller 460 is sent
`to the second network interface
`controller 480 of the second RAID controller 461.
`The first network interface controllers 470, 480 of two
`RAID controllers 460, 461 respectively supply data of the
`host computers 400 to 402 connected to the first hub 440 and
`the host computer 403 to 405 connected to the second hub
`441, and process information transmitted from the opposite
`network interface controllers 471, 481.
`If any one out of two RAID controllers 460, 461 has an
`occurrence of an error, the RAID controller having the error
`occurrence is removed from the network, and a second
`network interface controller of an opposite RAID controller
`not having the error occurrence takes over a function of a
`first network interface controller of the RAID controller
`
`having the error occurrence.
`FIG. 5 is a block diagram providing one embodiment of
`the host interface system as an external installation system
`between the RAID and the host computers in the present
`invention.
`
`As shown in FIG. 4, the present invention can be con-
`structed by a method of internally installing the hubs 440,
`441 in the RAID 490, and as shown in FIG. 5, the host
`computers 500, 501, 502, 503, 504 and 505 are connected to
`the RAID 530 by using external hubs 510 and 520.
`FIG. 6 is a block diagram showing one embodiment of the
`host
`interface system as a network switch between the
`inventive RAID and host computers.
`As shown in the drawing, a plurality of host computers
`600, 601, 602, 604 and 605 are connected to RAID through
`a network switch 610. In other words, information from a
`second network interface controller 622 of a first RAID
`controller 620 is sent to a first network interface controller
`632 of a second RAID controller 630, and information from
`a second network interface controller 632 of the second
`RAID controller 630 is transmitted to a first network inter-
`face controller 621 of the first RAID controller 620. Further,
`information from the first network interface controller 631
`of the second RAID controller 630 is transmitted to the
`second network interface controller 622 of the first RAID
`controller 620. Also,
`information from the first network
`interface controller 621 of the first RAID controller 620 is
`sent to the second network interface controller 632 of the
`second RAID controller 630.
`
`10
`
`15
`
`20
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`there is no distinction between respective ports,
`Just,
`representatively 611, of a network switch 610 at all and also,
`the internal structure of a network switch 610 can be
`
`60
`
`65
`
`configured according to a selection of a user (not shown in
`FIG. 6).
`In accordance with the present invention, as afore-men-
`tioned, even in a case of an error occurrence in a RAID
`controller, there exist two independent networks and two
`network interface controllers, and the bandwidth of a single
`network can be twice maintained. Accordingly, a function of
`fault tolerance between two RAID controllers can be con-
`
`structed without a drop of the bandwidth.
`It will be apparent to those skilled in the art that various
`modifications and variations can be made in the present
`
`DHPN-1001 / Page 9 0f10
`
`
`
`US 6,978,346 B2
`
`5
`invention without deviating from the spirit or scope of the
`invention. Thus,
`it is intended that the present invention
`cover the modifications and variations of this invention
`
`provided they come within the scope of the appended claims
`and their equivalents.
`What is claimed is:
`
`1. An apparatus for a redundant interconnection between
`multiple hosts and a RAID, comprising:
`a first RAID controlling units and a second RAID con-
`trolling unit for processing a requirement of numerous
`host computers, the first RAID controlling unit includ-
`ing a first network controlling unit and a second net-
`work controlling unit, and the second RAID controlling
`unit including a third network controlling unit and a
`fourth network controlling unit; and
`a plurality of connection units for connecting the first
`RAID controlling units and the second RAID control-
`ling unit to the numerous host computers, wherein the
`first RAID controlling unit and the second RAID
`controlling unit directly exchange information with the
`numerous host computers through the plurality of con-
`necting units, and the first network controlling unit
`exchanges information with the fourth network con-
`trolling unit, and the second network controlling unit
`exchanges information with the third network control-
`ling unit.
`2. The apparatus as recited in claim 1, wherein said
`respective RAID controlling units are connected to the
`plurality of individual connecting units.
`3. The apparatus as recited in claim 2, wherein the first
`network interface controlling unit is coupled to the connect-
`ing unit of one side and the second network interface
`controlling unit is coupled to the connecting unit of another
`side.
`
`4. The apparatus as recited in claim 3, wherein
`the first network interface controlling unit and the third
`network interface controlling unit process the require-
`ment of the numerous host computers; and
`the second network interface controlling unit and the
`fourth network controlling unit are used for communi-
`cation between the first RAID controlling unit and the
`second RAID controlling unit when the first and second
`RAID controlling units are not faulty and the second
`network interface controlling unit and the fourth net-
`work controlling unit are used for executing a function
`of the first network interface controlling unit and the
`third network controlling unit when one of the first
`RAID controlling unit and the second RAID control-
`ling unit is faulty.
`5. The apparatus as recited in claim 1, wherein said
`plurality of connecting units have at least three connection
`ports, two of the at least three connection ports is coupled to
`one of the first network interface controlling unit and the
`third network controlling unit and the rest of the connection
`ports being provided as a hub equipment connected with the
`numerous host computers.
`6. The apparatus as recited in claim 1, wherein said
`plurality of connecting units have at least three connection
`ports, two of the at least three connection port are coupled
`to one of the first network interface controlling unit and the
`third network controlling unit and the rest of the connection
`
`6
`ports being provided as a network switch equipment con-
`nected with the numerous host computers.
`7. The apparatus as recited in claim 1, wherein said
`plurality of connecting units have at least five connection
`ports, four of the at least five connection ports is coupled to
`one of the first network interface controlling unit and the
`third network controlling unit and the rest of the connection
`ports being provided as a switch connected with the numer-
`ous host computers.
`8. The apparatus as recited in claim 1, wherein the first
`network interface controlling unit of the first RAID control-
`ling unit being connected to a first connecting unit, the
`second network interface controlling unit of said first RAID
`controlling unit being connected to a second connecting
`unit,
`the third network interface controlling unit of the
`second RAID controlling unit being connected to the second
`connecting unit, and the fourth network interface controlling
`unit of the second RAID controlling unit being connected to
`the first connecting unit.
`9. An apparatus for a redundant interconnection between
`multiple host computers and a RAID, the apparatus com-
`prising:
`a plurality of connection units for connecting the host
`computers and the RAID;
`a first and a second RAID controllers, included in the
`RAID, each of which having a first network interface
`controller and a second network interface controller for
`
`processing requests from the plurality of the host
`computers connected through the plurality of the con-
`nection units,
`wherein the first network interface controller in the first
`
`RAID controller supplies data to the host computers
`connected through the plurality of connection units and
`processes information transmitted from the second net-
`work interface controller in the second RAID control-
`ler,
`wherein the first network interface controller in the sec-
`
`ond RAID controller supplies data to the host comput-
`ers connected through the plurality of connection units
`and processes information transmitted from the second
`network interface controller in the first RAID control-
`ler,
`wherein the second network interface controller in the
`
`first RAID controller is used for fault tolerance by
`performing functions of the first network interface
`controller in the second RAID controller when the
`
`second RAID controller is faulty, and
`wherein the second network interface controller in the
`second RAID controller is used for fault tolerance by
`performing functions of the first network interface
`controller in the first RAID controller when the first
`
`RAID controller is faulty, and
`in the first
`wherein the first network controlling unit
`RAID controlling unit exchanges information with the
`second network controlling unit in the second RAID
`controlling unit, and the second network controlling
`unit in the first RAID controlling unit exchanges infor-
`mation with the first network controlling unit in the
`second RAID controlling unit.
`*
`*
`*
`*
`
`*
`
`10
`
`15
`
`20
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`DHPN-1001 /Page10 0f10
`
`