公开(公告)号：US20130239121A1

公开(公告)日：2013-09-12

申请号：US13988249

申请日：2011-11-18

申请人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

发明人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

IPC分类号： G06F9/54

CPC分类号： G06F9/54 ,  G06F15/17337 ,  G06F15/17343 ,  G06F15/17356

摘要： A network architecture is used for the communication between elementary calculus units or nodes of a supercomputer to execute a super-calculus processing application, partitionable and scalable at the level of calculus power in the range of PetaFLOPS. The supercomputer comprises a plurality of modular structures, each of which comprises a plurality of elementary calculus units or nodes defined by node cards, a backplane, a root card, and a node communication network of the switched fabric fat tree type; ii) a synchronization architecture comprising a plurality of distinct node communication networks, configured for the communication of specific synchronization information different from network to network and with different characteristics; iii) a re-configurable Programmable Network Processor that implements the nodes both of the n-toroidal network and those of the synchronization networks.The node communication networks of the n-toroidal type and of the switched fabric fat tree type can be used alternately or simultaneously for the transmission between the calculus nodes of the same type of data and information also in a configuration of system partitions in order to achieve the desired interconnection topology of the nodes.

摘要翻译： 网络架构用于超级计算机的基本演算单元或节点之间的通信，以执行超演算处理应用，可在PetaFLOPS范围内的演算能级别进行可分割和可缩放。 超级计算机包括多个模块化结构，每个模块结构包括由交换结构胖树类型的节点卡，背板，根卡和节点通信网络定义的多个基本演算单元或节点; ii）包括多个不同节点通信网络的同步架构，被配置用于通信不同于网络到不同特征的特定同步信息; iii）可重新配置的可编程网络处理器，其实现n环形网络和同步网络的节点。 n环型和交换矩阵胖树类型的节点通信网络可以交替地或同时地用于在同一类型的数据的微积分节点之间的传输以及在系统分区的配置中的信息，以便实现 所需节点的互连拓扑。

公开(公告)号：US20130128460A1

公开(公告)日：2013-05-23

申请号：US13813192

申请日：2011-07-29

申请人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

发明人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

IPC分类号： H05K7/20

CPC分类号： H05K7/20254 ,  H05K7/20636 ,  H05K7/20772

摘要： Cooling device with liquid for electronic cards, in particular for high-performance processing units, comprising at least a hydraulic circuit in which a heat-carrying fluid flows in order to extract the heat produced by electronic components and/or hot spots present on an associated electronic card. The device comprises a cooling plate mechanically coupled to the electronic card so as to be inserted in a containing rack of the processing unit. The plate has a heat extraction surface facing and partly in contact, or at least in close proximity, with the electronic components and/or the hot spots. The hydraulic circuit is made in the thickness of the plate and has a geometric grid development, along which a plurality of hydraulic sub-circuits are disposed and switching means to define desired paths. The plate has sliding guide means able to cooperate with corresponding alignment means of the containing rack.

摘要翻译： 具有用于电子卡片的液体的冷却装置，特别是用于高性能处理装置的液体装置，至少包括一个液压回路，其中导热流体流动，以便提取由电子部件产生的热量和/或相关联的热点 电子卡。 该装置包括机械地联接到电子卡片以便插入处理单元的容纳架中的冷却板。 该板具有面对并部分地与电子部件和/或热点接触或至少紧密接近的热提取表面。 液压回路以板的厚度制成，并且具有几何栅格显影，多个液压子电路沿着该几何形状设置，切换装置限定所需的路径。 该板具有能够与容纳齿条的对应对准装置配合的滑动导向装置。

公开(公告)号：US09253920B2

公开(公告)日：2016-02-02

申请号：US13813192

申请日：2011-07-29

申请人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

发明人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

IPC分类号： H05K7/20

CPC分类号： H05K7/20254 ,  H05K7/20636 ,  H05K7/20772

摘要： Cooling device with liquid for electronic cards, in particular for high-performance processing units, comprising at least a hydraulic circuit in which a heat-carrying fluid flows in order to extract the heat produced by electronic components and/or hot spots present on an associated electronic card. The device comprises a cooling plate mechanically coupled to the electronic card so as to be inserted in a containing rack of the processing unit. The plate has a heat extraction surface facing and partly in contact, or at least in close proximity, with the electronic components and/or the hot spots. The hydraulic circuit is made in the thickness of the plate and has a geometric grid development, along which a plurality of hydraulic sub-circuits are disposed and switching means to define desired paths. The plate has sliding guide means able to cooperate with corresponding alignment means of the containing rack.

摘要翻译： 具有用于电子卡片的液体的冷却装置，特别是用于高性能处理装置的液体装置，至少包括一个液压回路，其中导热流体流动，以便提取由电子部件产生的热量和/或相关联的热点 电子卡。 该装置包括机械地联接到电子卡片以便插入处理单元的容纳架中的冷却板。 该板具有面对并部分地与电子部件和/或热点接触或至少紧密接近的热提取表面。 液压回路以板的厚度制成，并且具有几何栅格显影，多个液压子电路沿着该几何形状设置，切换装置限定所需的路径。 该板具有能够与容纳齿条的对应对准装置配合的滑动导向装置。

公开(公告)号：US09025595B2

公开(公告)日：2015-05-05

申请号：US13988249

申请日：2011-11-18

申请人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

发明人： Mauro Rossi ,  Giampietro Tecchiolli ,  Pierfrancesco Zuccato

IPC分类号： G06F9/54 ,  G06F9/46 ,  G06F15/173

CPC分类号： G06F9/54 ,  G06F15/17337 ,  G06F15/17343 ,  G06F15/17356

摘要： A network architecture is used for the communication between elementary calculus units or nodes of a supercomputer to execute a super-calculus processing application, partitionable and scalable at the level of calculus power in the range of PetaFLOPS. The supercomputer comprises a plurality of modular structures, each of which comprises a plurality of elementary calculus units or nodes defined by node cards a backplane, a root card, and a node communication network of the switched fabric fat tree type; ii) a synchronization architecture comprising a plurality of distinct node communication networks, configured for the communication of specific synchronization information different from network to network and with different characteristics; iii) a re-configurable Programmable Network Processor that implements the nodes both of the n-toroidal network and those of the synchronization networks.The node communication networks of the n-toroidal type and of the switched fabric fat tree type can be used alternately or simultaneously for the transmission between the calculus nodes of the same type of data and information also in a configuration of system partitions in order to achieve the desired interconnection topology of the nodes.

摘要翻译： 网络架构用于超级计算机的基本演算单元或节点之间的通信，以执行超演算处理应用，可在PetaFLOPS范围内的演算能级别进行可分割和可缩放。 超级计算机包括多个模块化结构，每个模块结构包括由节点卡定义的多个基本演算单元或节点，所述节点卡是交换结构胖树类型的背板，根卡和节点通信网络; ii）包括多个不同节点通信网络的同步架构，被配置用于通信不同于网络到不同特征的特定同步信息; iii）可重新配置的可编程网络处理器，其实现n环形网络和同步网络的节点。 n环型和交换矩阵胖树类型的节点通信网络可以交替地或同时地用于在同一类型的数据的微积分节点之间的传输以及在系统分区的配置中的信息，以便实现 所需节点的互连拓扑。