TW201823918A - Server system - Google Patents
Server system Download PDFInfo
- Publication number
- TW201823918A TW201823918A TW105143372A TW105143372A TW201823918A TW 201823918 A TW201823918 A TW 201823918A TW 105143372 A TW105143372 A TW 105143372A TW 105143372 A TW105143372 A TW 105143372A TW 201823918 A TW201823918 A TW 201823918A
- Authority
- TW
- Taiwan
- Prior art keywords
- signal
- host
- rack
- server system
- devices
- Prior art date
Links
Abstract
Description
[0001 ] 本發明涉及伺服器的技術領域，尤其涉及一種伺服器系統。[0001] The present invention relates to the technical field of servers, and in particular, to a server system.
[0002 ] 一般來說，傳統機架式伺服器在計算、儲存、網路資源配比上較為均衡，且具有很好的通用性。並且，在伺服器的發展歷程中，機架式伺服器承擔著非常重要的作用。 [0003 ] 然而，隨著互聯網公司單體業務的規模膨脹，傳統機架式伺服器在運行某一種資源密集型的應用時，例如大量的儲存磁盤的應用或高性能計算的應用，由於傳統機架式伺服器在計算、儲存、網路資源配比上無側重點，使得機架式伺服器無法有效負荷前述資源密集型的應用，而降低機架式伺服器的工作效率。為了有效地負荷資源密集型的應用，當傳統機架式伺服器的部署需要大量的儲存磁盤的應用時，就需要定制儲存型伺服器；當部署需要高性能計算的應用時，就需要定制計算型伺服器。如此，不僅將需要定制多種型號的機架式伺服器機型，造成成本消耗巨大，且無論是定制儲存性伺服器還是定制計算性伺服器都會另行定制整個機架內的構造，適用範圍低，不利於推廣。因此，伺服器的設計上仍有改善的空間。[0002] In general, the traditional rack server is more balanced in computing, storage, and network resource ratio, and has good versatility. Moreover, in the development of the server, the rack server plays a very important role. [0003] However, as the size of the Internet company's single business expands, traditional rack servers run a certain resource-intensive application, such as a large number of storage disk applications or high-performance computing applications, due to traditional machines The rack server has no focus on computing, storage, and network resource ratio, making the rack server unable to effectively load the aforementioned resource-intensive applications and reducing the efficiency of the rack server. In order to effectively load resource-intensive applications, when a traditional rack server deployment requires a large number of storage disk applications, a custom storage server is required; when deploying applications that require high performance computing, custom calculations are required. Type server. In this way, not only will the need to customize a variety of rack-type servo models, resulting in huge cost, and whether the custom storage server or custom computing server will be customized separately in the entire rack, the scope of application is low. Not conducive to promotion. Therefore, there is still room for improvement in the design of the server.
[0004 ] 本發明的主要目的在於提供一種伺服器系統，以解決現有技術存在的為了有效地負荷資源密集型的應用而需要設計多種伺服器類型，不利於推廣且適用範圍低的問題，進而有效地提升伺服器系統的運作效能。 [0005 ] 為解決上述問題，本發明實施例提供一種伺服器系統，包括：機架、機架管理控制器、多個主機裝置、多個儲存裝置與二信號交換器。機架管理控制器產生控制信號。多個主機裝置位於所述機架內。多個儲存裝置分別電性連接所述多個主機裝置，所述多個儲存裝置設置於所述機架內且位於所述多個主機裝置的下方。二信號交換器分別電性連接所述多個主機裝置及所述機架管理控制器，且每一所述信號交換器電性連接所述多個儲存裝置，所述二信號交換器設置於所述機架內且位於所述多個主機裝置的上方。其中，每一所述主機裝置通過所述二信號交換器接收所述控制信號以與所述多個儲存裝置其中至少一進行配對，使得每一所述主機裝置對與其配對的儲存裝置的資料進行存取與處理操作。 [0006 ] 所述二信號交換器各自包括溫度暫存器，所述機架管理控制器與所述溫度暫存器電性連接以自所述溫度暫存器接收機箱溫度資訊，並根據所述主機殼溫度資訊產生風扇轉速調整信號，所述風扇轉速調整信號通過所述二信號交換器傳輸至所述伺服器系統的風扇從而調整所述風扇的轉速。 [0007 ] 所述機架管理控制器接收工作模式調整指令，並依據所述工作模式調整指令產生工作模式調整信號給所述二信號交換器以使所述二信號交換器各自根據所述調整信號變更其工作模式設定資料。 [0008 ] 所述二信號交換器各包括電子抹除式可複寫唯讀記憶體，所述工作模式設定資料儲存於所述電子抹除式可複寫唯讀記憶體內。 [0009 ] 所述機架管理控制器接收軟體更新指令，並依據所述軟體更新指令，對所述二信號交換器進行軟體更新操作。 [0010 ] 所述機架管理控制器偵測所述二信號交換器的工作狀態，當所述二信號交換器的其中之一的工作狀態為斷線狀態時，工作狀態為斷線的所述信號交換機所配對的所述主機裝置與所述儲存裝置通過所述機架管理控制器重新分配至工作狀態為未斷線的所述信號交換機以與未斷線的所述信號交換器進行重新配對，使得重新配對的所述主機裝置與所述儲存裝置恢復正常工作。 [0011 ] 所述主機裝置包括主機及主機匯流排適配器，所述主機通過所述主機匯流排適配器電性連接所述儲存裝置及所述二信號交換器。 [0012 ] 所述二信號交換器通過SCSI接線與所述多個主機裝置和所述多個儲存裝置電性連接。 [0013 ] 根據本發明的技術方案，通過通過二信號交換器分別與機架管理控制器、多個主機裝置及多個儲存裝置電性連接，並且二信號交換器根據機架管理控制器所產生的控制信號，控制主機裝置與儲存裝置的配對關係，以便於根據不同的業務特點，靈活、彈性的調整主機裝置與其配對的儲存裝置，進而有效地提升伺服器系統的運作效能，提高伺服器系統的適用範圍。[0004] A main object of the present invention is to provide a server system to solve the problem that the prior art needs to design a plurality of server types in order to effectively load resource-intensive applications, which is disadvantageous for promotion and low application range, and is effective. Improve the operational efficiency of the server system. [0005] In order to solve the above problem, an embodiment of the present invention provides a server system, including: a rack, a rack management controller, a plurality of host devices, a plurality of storage devices, and two signal switches. The rack management controller generates control signals. A plurality of host devices are located within the rack. The plurality of storage devices are electrically connected to the plurality of host devices, and the plurality of storage devices are disposed in the rack and located below the plurality of host devices. The two signal exchanges are electrically connected to the plurality of host devices and the rack management controller, and each of the signal switches is electrically connected to the plurality of storage devices, and the two signal switches are disposed at the Within the rack and above the plurality of host devices. Each of the host devices receives the control signal through the two signal converters to pair with at least one of the plurality of storage devices, such that each of the host devices performs data on a storage device paired with the host device Access and processing operations. [0006] the two signal converters each include a temperature register, the rack management controller is electrically connected to the temperature register to receive chassis temperature information from the temperature register, and according to the The main cabinet temperature information generates a fan speed adjustment signal, and the fan speed adjustment signal is transmitted to the fan of the server system through the two signal exchangers to adjust the rotation speed of the fan. [0007] the rack management controller receives an operation mode adjustment instruction, and generates an operation mode adjustment signal to the two signal exchangers according to the operation mode adjustment command, so that the two signal converters respectively according to the adjustment signal Change the working mode setting data. [0008] The two signal converters each comprise an electronic erasable rewritable read-only memory, and the working mode setting data is stored in the electronic erasable rewritable read-only memory. [0009] The rack management controller receives a software update instruction, and performs a software update operation on the two signal switches according to the software update instruction. [0010] the rack management controller detects an operating state of the two signal exchangers, and when the working state of one of the two signal converters is a disconnected state, the working state is the disconnected state The host device paired with the signal switch and the storage device are reassigned by the rack management controller to the signal switch that is in an uninterrupted state to re-pair with the unswitched signal exchanger The re-paired host device and the storage device resume normal operation. [0011] The host device includes a host and a host busbar adapter, and the host is electrically connected to the storage device and the two signal exchangers through the host busbar adapter. [0012] The two signal exchangers are electrically connected to the plurality of host devices and the plurality of storage devices through SCSI wires. [0013] According to the technical solution of the present invention, the two are separately electrically connected to the rack management controller, the plurality of host devices, and the plurality of storage devices, and the two signal switches are generated according to the rack management controller. The control signal controls the pairing relationship between the host device and the storage device, so as to flexibly and flexibly adjust the host device and its paired storage device according to different service characteristics, thereby effectively improving the operating efficiency of the server system and improving the server system. Scope of application.
[0015 ] 以下將配合圖式及實施例來詳細說明本發明之實施方式，藉此對本發明如何應用技術手段來解決技術問題並達成技術功效的實現過程能充分理解並據以實施。 [0016 ] 本發明通過二信號交換器分別與機架管理控制器、多個主機裝置及多個儲存裝置電性連接，並且二信號交換器根據機架管理控制器所產生的控制信號，控制主機裝置與儲存裝置的配對關係，以便於根據不同的業務特點，靈活、彈性的調整主機裝置與其配對的儲存裝置，進而有效地提升伺服器系統的運作效能。 [0017 ] 為使本發明的目的、技術方案和優點更加清楚，以下結合附圖及具體實施例，對本發明作進一步地詳細說明。 [0018 ] 根據本發明的實施例，提供了一種伺服器系統。參照本發明之第1圖及第2圖所示內容。伺服器系統100包括：機架110、機架管理控制器120、多個主機裝置130_1~130_N、多個儲存裝置140_1~140_N與二信號交換器150、160，其中N分別為大於1的正整數。 [0019 ] 機架管理控制器（Rack Management Controller, RMC）120產生控制信號，以控制信號交換器150、160、主機裝置130_1~130_N與儲存裝置140_1~140_N之間的運作。 [0020 ] 多個主機裝置130_1~130_N位於所述機架110內。進一步來說，主機裝置130_1~130_N由上往下依序配置於所述機架110內，且主機裝置130_1~130_N彼此平行配置。 [0021 ] 多個儲存裝置140_1~140_N分別電性連接所述主機裝置130_1~130_N，所述多個儲存裝置140_1~140_N設置於所述機架110內且位於所述主機裝置130_1~130_N的下方。進一步來說，儲存裝置140_1~140_N例如設置位於主機裝置130_N的下方，即機架110的下方。在本實施例中，所述多個儲存裝置140_1~140_N各自為SAS JBOD儲存裝置。 [0022 ] 二信號交換器150、160分別電性連接所述多個主機裝置130_1~130_N及所述機架管理控制器120，且每一所述信號交換器150、160電性連接所述多個儲存裝置140_1~140_N，所述二信號交換器150、160設置於所述機架110內且位於所述主機裝置130_1~130_N的上方，即機架110內部的上方。在本實施例中，所述二信號交換器150、160例如為SAS交換器。 [0023 ] 每一所述主機裝置130_1~130_N通過所述二信號交換器150、160接收所述控制信號以與所述多個儲存裝置140_1~140_N其中至少一進行配對，使得每一所述主機裝置130_1~130_N對與其配對的儲存裝置的資料進行存取與處理操作。 [0024 ] 舉例來說，在一些實施例中，機架管理控制器120產生的控制信號包括主機裝置130_1與儲存裝置140_1、140_2配對、主機裝置130_2、130_3與儲存裝置140_3配對、主機裝置130_4與儲存裝置140_4、140_5、140_6配對，並且發送此控制信號給信號交換器150。信號交換器150接收到上述控制信號並傳輸至主機裝置130_1、130_2、130_3、130_4，主機裝置130_1根據控制信號搜尋儲存裝置140_1、140_2進行配對、主機裝置130_2、130_3根據控制信號搜尋儲存裝置140_3進行配對，主機裝置130_4根據控制信號搜尋儲存裝置140_4、140_5、140_6進行配對。 [0025 ] 接著，主機裝置130_1可對與其配對的儲存裝置（即儲存裝置140_1、140_2）的資料進行存取與處理操作。主機裝置130_2、130_3可對與其配對的儲存裝置（即儲存裝置140_3）的資料進行存取與處理操作。主機裝置130_4可對與其配對的儲存裝置（即儲存裝置140_4、140_5、140_6）的資料進行存取與處理操作。 [0026 ] 在一具體實施例中，在機架110中配置主機裝置130_1、130_2及儲存裝置140_1、140_2……140_12， 機架管理控制器120產生控制信號並經由信號交換器150、160傳輸控制信號至主機裝置130_1、130_2，主機裝置130_1根據控制資訊與6個儲存裝置140_1、140_2……140_6配對以與該些儲存裝置進行資料的存取與處理操作，主機裝置130_2根據控制資訊與6個儲存裝置140_7、140_8……140_12配對以與該些儲存裝置進行資料的存取與處理操作。 [0027 ] 上述的實施例僅為本發明的一種實施方式，並不用以限定本發明，且用戶可視其需求（例如大量的儲存磁盤的應用需求或高性能計算的應用需求），調整主機裝置130_1~130_N與儲存裝置140_1~140_N的配置關係及配置數量。而主機裝置130_1~130_N與儲存裝置140_1~140_N的其他配對關係的實施方式可參考前述實施例，故在此不再贅述。 [0028 ] 由上述的說明可知，通過機架管理控制器120所產生的控制信號，可以控制信號交換器150、160來調整主機裝置130_1~130_N與儲存裝置140_1~140_N的配對關係，以便於根據不同的業務特點，靈活、彈性的調整主機裝置與其配對的儲存裝置。也就是說，當對計算性能和訪問延遲有很高要求時，可以相應增加主機裝置的數量，減少儲存裝置的數量；當對磁盤的儲存容量有需求時，可以相應減少主機裝置的數量，同時增加儲存裝置的數量。如此一來，將可以有效地提升伺服器系統100的運作效能。 [0029 ] 另外，參照第1圖所示，所述二信號交換器150、160可各自包括溫度暫存器151、161。所述機架管理控制器120與所述溫度暫存器151、161電性連接，以自所述溫度暫存器151、161接收機箱溫度資訊，並根據所述主機殼溫度資訊產生風扇轉速調整信號。接著，所述機架管理控制器120將風扇轉速調整信號通過所述二信號交換器151、161輸出至所述伺服器系統100的風扇，從而調整所述風扇的轉速。 [0030 ] 也就是說，信號交換器150、160會監控機架110內的主機殼溫度，並將主機殼溫度存儲於溫度暫存器151、161。機架管理控制器120可讀取溫度暫存器151、161中的主機殼溫度資訊，以根據主機殼溫度資訊產生對應的風扇速度調整信號。舉例來說，當機架管理控制器120判斷主機殼溫度資訊中的溫度高於溫度閾值，則產生例如增加風扇轉速的風扇轉速調整信號。接著，機架管理控制器120將增加風扇轉速的風扇轉速調整信號輸出伺服器系統100的風扇，使得風扇的轉速增加，以便於有效地降低機架110內的溫度。 [0031 ] 當機架管理控制器120判斷主機殼溫度資訊中的溫度等於或低於溫度閾值，則產生例如降低風扇轉速的風扇轉速調整信號。接著，機架管理控制器120將降低風扇轉速的風扇轉速調整信號輸出伺服器系統100的風扇，使得風扇的轉速降低，以便於有效地降低伺服器系統100的耗電量。 [0032 ] 此外，所述機架管理控制器120還可接收工作模式調整指令，並依據所述工作模式調整指令產生工作模式調整信號給所述二信號交換器150、160以使所述二信號交換器150、160各自根據所述調整信號變更其工作模式設定資料。並且，所述二信號交換器150、160各包括電子抹除式可複寫唯讀記憶體（Electrically Erasable Programmable Read-Only Memory, EEPROM）152、162，所述工作模式設定資料儲存於所述電子抹除式可複寫唯讀記憶體152、162內。 [0033 ] 也就是說，維護工程師可以遠端發出工作模式調整指令給機架管理控制器120，使得機架管理控制器120對應產生工作模式調整信號給所述二信號交換器150、160，以調整信號交換器150、160工作模式設定。並且，所述機架管理控制器120還可以接收例如由維護工程師所發出的軟體更新指令，並依據所述軟體更新指令，對所述二信號交換器150、160進行軟體更新操作。如此一來，可有效地增加伺服器系統100的工作效率，並增加使用上的便利性。 [0034 ] 進一步的，所述機架管理控制器120可偵測所述二信號交換器150、160的工作狀態，以便於根據信號交換器150、160的工作狀態，調整伺服器系統100的運作。當所述二信號交換器150、160的其中之一（例如信號交換器150）的工作狀態為斷線狀態時，工作狀態為斷線的所述信號交換機（例如信號交換器150）所配對的所述主機裝置與所述儲存裝置通過所述機架管理控制器120重新分配至工作狀態為未斷線的所述信號交換機（例如信號交換器160）以與未斷線的所述信號交換器（信號交換器160）進行重新配對，使得重新配對的所述主機裝置與所述儲存裝置恢復正常工作。 [0035 ] 在一些實施例中，當信號交換器150的工作狀態為斷線狀態時，主機裝置130_1、130_2、130_3、130_4與儲存裝置140_1、140_2、140_3、140_4、140_5、140_6可通過機架管理控制器120重新分配至工作狀態為未斷線的信號交換機160，以便由信號交換機160替代信號交換機150傳遞控制信號，使得主機裝置130_1根據信號交換機160所傳輸的控制信號與儲存裝置140_1、140_2進行配對、主機裝置130_2、130_3與儲存裝置140_3進行配對，以及主機裝置130_4同理與儲存裝置140_4、140_5、140_6進行配對，從而使得重新配對的主機裝置130_1與儲存裝置140_1、140_2恢復正常工作、重新配對的主機裝置130_2、130_3與儲存裝置140_3恢復正常工作及重新配對的主機裝置130_4與儲存裝置140_4、140_5、140_6恢復正常工作。如此一來，伺服器系統100不會因為二信號交換器150、160的其中之一的工作狀態為斷線狀態時而影響其工作效率，且仍可正常運作，將可增加伺服器系統100的工作效率並增加使用上的便利性。 [0036 ] 進一步的，在本實施例中，所述主機裝置130_1~130_N例如包括各自主機131_1~131_N及主機匯流排適配器132_1~132_N。所述主機131_1~131_N通過所述主機匯流排適配器132_1~132_N電性連接所述儲存裝置140_1~140_N及所述二信號交換器150、160。並且，所述二信號交換器150、160通過SCSI接線與所述多個主機裝置130_1~130_N和所述多個儲存裝置140_1~140_N電性連接。 [0037 ] 綜上所述，根據本發明的技術方案，通過二信號交換器分別與機架管理控制器、多個主機裝置及多個儲存裝置電性連接，並且二信號交換器根據機架管理控制器所產生的控制信號，控制主機裝置與儲存裝置的配對關係，以便於根據不同的業務特點，靈活、彈性的調整主機裝置與其配對的儲存裝置，擴大伺服器系統的適用範圍，進而有效地提升伺服器系統的運作效能。另外，還可根據主機殼溫度資訊調整出伺服器系統的風扇的轉速，以便於有效地降低機架內的溫度或是降低伺服器系統的耗電量。此外，機架管理控制器還可偵測二信號交換器的工作狀態，以便於二信號交換器的其中之一的工作狀態為斷線狀態，將工作狀態為斷線的信號交換機所配對的主機裝置與儲存裝置通過未斷線的信號交換機進行重新配對，使得重新配對的所述主機裝置與所述儲存裝置恢復正常工作，如此可增加伺服器系統的工作效率並增加使用上的便利性。 [0038 ] 雖然本發明以前述之實施例揭露如上，然其並非用以限定本發明，任何熟習相像技藝者，在不脫離本發明之精神和範圍內，當可作些許之更動與潤飾，因此本發明之專利保護範圍須視本說明書所附之申請專利範圍所界定者為準。[0015] The embodiments of the present invention will be described in detail below with reference to the drawings and embodiments, so that the application of the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. [0016] The present invention is electrically connected to the rack management controller, the plurality of host devices, and the plurality of storage devices respectively by the two signal switches, and the two signal switches control the host according to the control signals generated by the rack management controller. The pairing relationship between the device and the storage device is to flexibly and flexibly adjust the host device and its paired storage device according to different service characteristics, thereby effectively improving the operating efficiency of the server system. [0017] The present invention will be further described in detail below with reference to the drawings and specific embodiments. [0018] According to an embodiment of the invention, a server system is provided. Reference is made to the contents shown in Figs. 1 and 2 of the present invention. The server system 100 includes a rack 110, a rack management controller 120, a plurality of host devices 130_1~130_N, a plurality of storage devices 140_1~140_N and two signal switches 150, 160, wherein N is a positive integer greater than one, respectively. . [0019] A Rack Management Controller (RMC) 120 generates control signals to control operations between the signal switches 150, 160, the host devices 130_1~130_N, and the storage devices 140_1~140_N. [0020] A plurality of host devices 130_1~130_N are located in the rack 110. Further, the host devices 130_1~130_N are sequentially disposed in the rack 110 from top to bottom, and the host devices 130_1~130_N are arranged in parallel with each other. [0021] The plurality of storage devices 140_1~140_N are electrically connected to the host devices 130_1~130_N, and the plurality of storage devices 140_1~140_N are disposed in the rack 110 and located below the host devices 130_1~130_N. . Further, the storage devices 140_1~140_N are disposed, for example, below the host device 130_N, that is, below the chassis 110. In this embodiment, each of the plurality of storage devices 140_1~140_N is a SAS JBOD storage device. [0022] The two signal switches 150 and 160 are electrically connected to the plurality of host devices 130_1~130_N and the rack management controller 120, and each of the signal switches 150 and 160 is electrically connected to the plurality of The storage devices 140_1~140_N are disposed in the rack 110 and above the host devices 130_1~130_N, that is, above the inside of the rack 110. In this embodiment, the two signal switches 150, 160 are, for example, SAS switches. [0023] each of the host devices 130_1~130_N receives the control signal through the two signal switches 150, 160 to pair with at least one of the plurality of storage devices 140_1~140_N, such that each of the hosts The devices 130_1~130_N access and process the data of the paired storage device. [0024] For example, in some embodiments, the control signals generated by the rack management controller 120 include the host device 130_1 paired with the storage devices 140_1, 140_2, the host devices 130_2, 130_3 paired with the storage device 140_3, and the host device 130_4 and The storage devices 140_4, 140_5, 140_6 are paired and send this control signal to the signal switch 150. The signal converter 150 receives the above control signals and transmits them to the host devices 130_1, 130_2, 130_3, and 130_4. The host device 130_1 searches for the storage devices 140_1 and 140_2 for pairing according to the control signals, and the host devices 130_2 and 130_3 search for the storage devices 140_3 according to the control signals. Pairing, the host device 130_4 searches for the storage devices 140_4, 140_5, 140_6 according to the control signals for pairing. [0025] Next, the host device 130_1 can perform access and processing operations on the data of the storage device (ie, the storage devices 140_1, 140_2) that it is paired with. The host devices 130_2, 130_3 can access and process the data of the storage device (ie, the storage device 140_3) with which they are paired. The host device 130_4 can access and process the data of the storage device (ie, the storage devices 140_4, 140_5, 140_6) with which it is paired. In a specific embodiment, the host devices 130_1, 130_2 and the storage devices 140_1, 140_2, ..., 140_12 are configured in the rack 110, and the rack management controller 120 generates control signals and transmits and controls the signals via the signal switches 150, 160. The signal is sent to the host device 130_1, 130_2, and the host device 130_1 is paired with the six storage devices 140_1, 140_2, ..., 140_6 according to the control information to perform data access and processing operations with the storage devices. The host device 130_2 is based on the control information and six The storage devices 140_7, 140_8, ..., 140_12 are paired to perform data access and processing operations with the storage devices. The above embodiments are only one embodiment of the present invention, and are not intended to limit the present invention, and the user may adjust the host device 130_1 according to the needs thereof (for example, application requirements of a large number of storage disks or application requirements of high performance computing). The configuration relationship and the number of configurations of ~130_N and storage devices 140_1~140_N. For the implementation of other matching relationships between the host devices 130_1~130_N and the storage devices 140_1~140_N, reference may be made to the foregoing embodiments, and details are not described herein. [0028] It can be seen from the above description that the signal converters 150 and 160 can be controlled by the control signals generated by the rack management controller 120 to adjust the pairing relationship between the host devices 130_1~130_N and the storage devices 140_1~140_N. Different business features, flexible and flexible adjustment of the host device and its paired storage device. That is to say, when there is a high requirement for computing performance and access delay, the number of host devices can be increased correspondingly, and the number of storage devices can be reduced; when there is a demand for the storage capacity of the disk, the number of host devices can be reduced correspondingly. Increase the number of storage devices. As a result, the operational performance of the server system 100 can be effectively improved. [0029] In addition, referring to FIG. 1, the two signal exchangers 150, 160 may each include a temperature register 151, 161. The rack management controller 120 is electrically connected to the temperature registers 151, 161 to receive chassis temperature information from the temperature registers 151, 161, and generate fan speed according to the host case temperature information. Adjust the signal. Next, the rack management controller 120 outputs a fan speed adjustment signal to the fan of the server system 100 through the two signal exchangers 151, 161, thereby adjusting the rotation speed of the fan. [0030] That is, the signal switches 150, 160 monitor the temperature of the main cabinet in the rack 110 and store the main cabinet temperature in the temperature registers 151, 161. The rack management controller 120 can read the host case temperature information in the temperature registers 151, 161 to generate a corresponding fan speed adjustment signal according to the host case temperature information. For example, when the rack management controller 120 determines that the temperature in the main cabinet temperature information is higher than the temperature threshold, a fan speed adjustment signal such as increasing the fan speed is generated. Next, the rack management controller 120 outputs a fan speed adjustment signal that increases the fan speed to the fan of the servo system 100 such that the speed of the fan is increased to effectively reduce the temperature within the rack 110. [0031] When the rack management controller 120 determines that the temperature in the main cabinet temperature information is equal to or lower than the temperature threshold, a fan speed adjustment signal such as a decrease in the fan speed is generated. Next, the rack management controller 120 outputs the fan speed adjustment signal that reduces the fan speed to the fan of the servo system 100, so that the rotation speed of the fan is lowered, so as to effectively reduce the power consumption of the server system 100. [0032] In addition, the rack management controller 120 may further receive an operation mode adjustment instruction, and generate an operation mode adjustment signal to the two signal switches 150, 160 according to the operation mode adjustment instruction to enable the two signals. The switches 150, 160 each change their operating mode setting data in accordance with the adjustment signal. Moreover, the two signal switches 150 and 160 each include an Electrically Erasable Programmable Read-Only Memory (EEPROM) 152, 162, and the working mode setting data is stored in the electronic wipe. The divide can be overwritten in the read only memory 152, 162. [0033] That is, the maintenance engineer can remotely issue the working mode adjustment command to the rack management controller 120, so that the rack management controller 120 correspondingly generates the working mode adjustment signal to the two signal switches 150, 160 to The signal switch 150, 160 operating mode settings are adjusted. Moreover, the rack management controller 120 can also receive, for example, a software update command issued by a maintenance engineer, and perform a software update operation on the two signal switches 150 and 160 according to the software update command. In this way, the working efficiency of the server system 100 can be effectively increased, and the convenience in use is increased. [0034] Further, the rack management controller 120 can detect the working states of the two signal switches 150, 160 to adjust the operation of the server system 100 according to the working states of the signal switches 150, 160. . When the operating state of one of the two signal exchangers 150, 160 (eg, the signal switch 150) is in a disconnected state, the signal switch (eg, the signal switch 150) operating in a disconnected state is paired The host device and the storage device are redistributed by the rack management controller 120 to the signal switch (eg, the signal switch 160) that is in an uninterrupted state to communicate with the unswitched signal exchanger The (signal exchanger 160) performs re-pairing so that the re-paired host device and the storage device resume normal operation. [0035] In some embodiments, when the operating state of the signal switch 150 is the disconnected state, the host devices 130_1, 130_2, 130_3, 130_4 and the storage devices 140_1, 140_2, 140_3, 140_4, 140_5, 140_6 may pass through the rack. The management controller 120 is redistributed to the signal switch 160 whose working state is unbroken, so that the signal switch 160 transmits the control signal instead of the signal switch 150, so that the host device 130_1 transmits the control signal and the storage device 140_1, 140_2 according to the signal switch 160. Pairing, the host devices 130_2, 130_3 are paired with the storage device 140_3, and the host device 130_4 is paired with the storage devices 140_4, 140_5, 140_6, so that the re-paired host device 130_1 and the storage devices 140_1, 140_2 resume normal operation, The re-paired host devices 130_2, 130_3 and the storage device 140_3 resume normal operation and the re-paired host device 130_4 and the storage devices 140_4, 140_5, 140_6 resume normal operation. In this way, the server system 100 does not affect the working efficiency when one of the two signal converters 150, 160 is in the disconnected state, and can still operate normally, and the server system 100 can be increased. Work efficiency and increase ease of use. [0036] Further, in this embodiment, the host devices 130_1~130_N include, for example, respective hosts 131_1~131_N and host busbar adapters 132_1~132_N. The hosts 131_1~131_N are electrically connected to the storage devices 140_1~140_N and the two signal switches 150, 160 through the host busbar adapters 132_1~132_N. Moreover, the two signal switches 150 and 160 are electrically connected to the plurality of host devices 130_1~130_N and the plurality of storage devices 140_1~140_N through SCSI wires. [0037] In summary, according to the technical solution of the present invention, the rack management controller, the plurality of host devices, and the plurality of storage devices are respectively electrically connected by the two signal switches, and the two signal switches are managed according to the rack. The control signal generated by the controller controls the pairing relationship between the host device and the storage device, so as to flexibly and flexibly adjust the host device and its paired storage device according to different service characteristics, thereby expanding the application range of the server system, thereby effectively Improve the performance of the server system. In addition, the speed of the fan of the server system can be adjusted according to the temperature information of the main body casing, so as to effectively reduce the temperature in the rack or reduce the power consumption of the server system. In addition, the rack management controller can also detect the working state of the two signal converters, so that one of the two signal converters is in a disconnected state, and the working state is a host of the disconnected signal switch. The device and the storage device are re-paired by the unbroken signal switch, so that the re-paired host device and the storage device resume normal operation, which can increase the working efficiency of the server system and increase the convenience in use. [0038] While the present invention has been described above in the foregoing embodiments, it is not intended to limit the present invention, and it is obvious that those skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. The patent protection scope of the present invention is defined by the scope of the patent application attached to the specification.
[0039 ][0039]
100‧‧‧伺服器系統100‧‧‧Server System
110‧‧‧機架110‧‧‧Rack
120‧‧‧機架管理控制器120‧‧‧Rack Management Controller
130_1~130_N‧‧‧主機裝置130_1~130_N‧‧‧Host device
131_1~131_N‧‧‧主機131_1~131_N‧‧‧Host
132_1~132_N‧‧‧主機匯流排適配器132_1~132_N‧‧‧Host Bus Adapter
140_1~140_N‧‧‧儲存裝置140_1~140_N‧‧‧Storage device
150、160‧‧‧信號交換器150, 160‧‧‧Signal Exchanger
151、161‧‧‧溫度暫存器151, 161‧‧ ‧ temperature register
152、162‧‧‧電子抹除式可複寫唯讀記憶體152, 162‧‧‧Electronic erasable rewritable read-only memory
[0014 ] 第1圖是根據本發明實施例的伺服器系統的電路圖。 第2圖是根據本發明實施例的伺服器系統的結構示意圖。[0014] FIG. 1 is a circuit diagram of a server system in accordance with an embodiment of the present invention. Fig. 2 is a block diagram showing the structure of a server system according to an embodiment of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105143372A TW201823918A (en) | 2016-12-27 | 2016-12-27 | Server system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105143372A TW201823918A (en) | 2016-12-27 | 2016-12-27 | Server system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TW201823918A true TW201823918A (en) | 2018-07-01 |
Family
ID=63640106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW105143372A TW201823918A (en) | 2016-12-27 | 2016-12-27 | Server system |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TW201823918A (en) |
-
2016
- 2016-12-27 TW TW105143372A patent/TW201823918A/en unknown
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11615044B2 (en) | Graphics processing unit peer-to-peer arrangements | |
| JP4814617B2 (en) | Storage system | |
| CN103117866B (en) | Switch fabric management method and system | |
| US8959374B2 (en) | Power management for devices in a data storage fabric | |
| US20050083853A1 (en) | System, method, and program for determining the availability of paths to a device | |
| US11086813B1 (en) | Modular non-volatile memory express storage appliance and method therefor | |
| JP2001256003A (en) | Disk array controller, its disk array control unit and its expanding method | |
| JP2012150623A (en) | Computer system and hba identifier takeover system for pci card | |
| US20150046646A1 (en) | Virtual Network Disk Architectures and Related Systems | |
| CN105446657A (en) | Method for monitoring RAID card | |
| CN108664440A (en) | Interface server and cabinet | |
| US11231858B2 (en) | Dynamically configuring a storage system to facilitate independent scaling of resources | |
| US6748557B1 (en) | System, method, and program for determining the availability of interfaces to a control unit | |
| CN106776452B (en) | Server system | |
| WO2011125106A1 (en) | Storage system configured from plurality of storage modules and method for switching coupling configuration of storage modules | |
| TW202020607A (en) | Hard disk module and server system | |
| US6745347B1 (en) | System, method and program for determining the availability of interfaces to a device from information provided by the device | |
| TW201823918A (en) | Server system | |
| US10649943B2 (en) | System and method for I/O aware processor configuration | |
| CN105446654A (en) | Server control method and rack controller | |
| CN105511990A (en) | Novel dual-redundancy storage control node framework based on fusion framework | |
| US20230078518A1 (en) | Systems and methods for collapsing resources used in cloud deployments | |
| WO2019043815A1 (en) | Storage system | |
| US10503673B2 (en) | Storage system and method of storage system for constructing system | |
| US20180285311A1 (en) | Server having ip miniature computing unit |