Abstract: Delivery of power to one or more servers is controlled by energizing a control circuit and transmitting, from the control circuit to a first power source, a first signal indicating that the control circuit is energized. After transmitting the first signal indicating that the control circuit is energized, the control circuit receives a second signal indicating that the first power source is ready to provide power. The control circuit also receives, from a server, a third signal indicating that the server is energized. Having received the second signal indicating that the first power source is ready to provide power and the third signal indicating that the server is energized, the control circuit transmit to the server a fourth signal indicating that the server may start receiving power from the first power source.

Abstract: A connection between a power distribution unit (PDU) and an electric device is identified by generating a data packet containing an identity of the electric device, transmitting the data packet by varying, on a power line connecting an output socket of the PDU to the electric device, a power level between distinct intensities for transmitting distinct logical values of the data packet, sensing, at the PDU, the power level on the power line, reporting, to an identification module, a socket number of the output socket connected to the power line and successive power level data obtained by sensing the power level on the power line, decoding the identity of the electric device by monitoring the successive power level data reported by the PDU, and storing, in a database, a reference between the identity of the electric device, an identity of the PDU, and the socket number.

Abstract: A method and system for classifying data packet fields are disclosed. They associate a final tag to each of the fields in a data packet in relation to a set of classifying rules, and involve building a decision tree using a recursive algorithm to apply the set of classifying rules on the data packet fields, mapping each node of the built decision tree respectively to a processing element of a FPGA, each processing element comprising a processor and a memory, pipelining all mapped processing elements, and processing the data packet fields through the pipelined and mapped processing elements.

Abstract: A liquid cooling device for cooling at least one target component includes a base member configured to be in thermal contact with the at least one target component to be cooled by the liquid cooling device. The base member defines a plurality of pockets spaced apart from one another. A plurality of cover members are connected to the base member and are at least partially received in a corresponding pocket. Each cover member has a fluid inlet and outlet. The cover and base members define together a plurality of fluid conduits that are independent from one another. Each fluid conduit is defined between a cover member and the base member such that a number of the fluid conduits is equal to a number of the cover members. Each fluid conduit respectively receives and discharges fluid through the fluid inlet and the fluid outlet of the corresponding cover member.

Abstract: A power supply combination comprises two multi-voltage power sources having respective power rails delivering power at a first voltage and respective power rails delivering power at a second voltage. The power supply combination also comprises two single-voltage power sources having respective power rails delivering power at the first voltage. A first power combining circuit is electrically connected to the first voltage power rails of two multi-voltage power sources and to the two single-voltage power sources and delivers power at the first voltage to a first voltage input of a load. A second power combining circuit is electrically connected to the second voltage power rails of the two multi-voltage power sources and delivers power at the second voltage to a second voltage input of the load. A power supply network may connect a plurality of power supply combinations to a power management unit for reporting failures of the power sources.

Abstract: A rack adapted for receiving one or more components is disclosed. The rack includes a backplane, a pair of side panels extending from the backplane and internal support members on each side to receive and mechanically guide an initial alignment of components upon their initial insertion in the rack. A pair of male connectors mounted to the backplane is configured to mate with a corresponding pair of female connectors of each component to mechanically guide a final alignment of each component when the component is further inserted in the rack. Mechanical guidance may also be provided by, or supplemented with, a connection capable of providing liquid cooling to the rack. A system including the rack and the component inserted in the rack is also disclosed.

Abstract: Methods and systems are described for connecting to network services on a private network. A gateway device may coordinate communications between a client device on a public network and a host device on a private network. The client device may request to access the host device via the gateway device. The gateway device may authenticate the client device. The gateway device may transcode communications between the client device and the host device, thereby masking the address of the client device and the host device. The gateway device may maintain two different encryption methods between the client device and the gateway device, and the gateway device and the host device.

Abstract: A health of a server infrastructure is supervised by sending measurement requests, from a polling node to a server of the server infrastructure, at each of successive polling periods. Measurements received from the server in response to the measurement requests are stored in a database. A machine learning system is trained using accumulated measurements stored in the database to construct a prediction model for the server. Compliance of a latest measurement received from the server with the prediction model for the server is evaluated. A corrective action is taken if the latest measurement does not comply with the prediction model for the server. Measurements received from a group of servers may be aggregated and the machine learning system may construct a prediction model for the group of servers. Compliance to the prediction model may be evaluated for the group of servers.

Abstract: A heat exchanger assembly includes: a frame; a heat exchanger panel mounted to the frame and configured to exchange heat with air flowing therethrough, the heat exchanger panel being disposed at an inclined orientation; a fan assembly disposed vertically above the heat exchanger panel; and a sound dampening device disposed within an interior space of the heat exchanger assembly such that air is pulled into the interior space through the heat exchanger panel and then flows through the sound dampening device before being discharged from the heat exchanger assembly via the fan assembly. The sound dampening device includes baffle members having sound absorbing material and spaced apart from one another for allowing air flow therebetween. Each baffle member extends at an angle relative to a plane extending through the upper and lower ends of the heat exchanger panel so as to direct air flow upwardly toward the fan assembly.

Abstract: A fluid circuit monitoring system for monitoring a fluid circuit includes a pressure sensor for sensing a pressure of fluid flowing in the fluid circuit and an interface member fluidly communicating the fluid circuit to the pressure sensor. The fluid circuit monitoring system may be used to monitor a pressure or temperature differential between an inlet and outlet of the fluid circuit and determine a fluid flow rate of the fluid circuit. The pressure differential, temperature differential, or fluid flow rate can be logged and used to determine if preventative maintenance should be performed on the fluid circuit.

Abstract: A data packet received at a service infrastructure is screened by a data packet cleaning system that successively applies each signature of a set of signatures as a mask to a predetermined area of a content of the data packet. If there is a byte-for-byte match between the predetermined area of the content and one of the signatures, an action corresponding to the matched signature is taken. The action is selected from unconditionally forwarding the data packet toward a server of the service infrastructure, unconditionally discarding the data packet, forwarding the data packet toward the server of the service infrastructure if a current flow of data packets being forwarded to the server is less than a flow threshold, and discarding the data packet if the current flow of data packets being forwarded to the server meets or exceeds the flow threshold.

Abstract: Disaggregated storage clusters are disclosed. These disaggregated storage clusters include a plurality of storage targets coupled to each other through a switch and including storage targets including storage and data services storage targets. Data and requests can for storage areas maintained by the storage cluster can be routed between the target of the storage clusters based on pipeline definitions for those storage areas.

Abstract: A cooling device comprises a first casing for direct mounting on a heat-generating component and a second casing mounted on the first casing. One casing includes an internal channel connected to a cold inlet and to a hot outlet allowing a heat-transfer fluid to flow in the internal channel. The other casing includes a storage containing a phase change material (PCM) changing from a solid state to a liquid state to transfer thermal energy from the heat-generating unit to the PCM, the PCM changing from the liquid state to the solid state to transfer thermal energy from the PCM to the heat-transfer fluid. The cooling device may be integrated in a cooling circuit of a cooling system including an external cooling unit, or in a closed loop cooling circuit of a cooling arrangement that transfers heat from the closed loop to an open loop.

Abstract: A heat exchanger stack includes a rack structure and heat exchangers disposed above one another and supported thereby. Each heat exchanger includes: a frame; a tubing arrangement configured to circulate fluid therein; a plurality of fins; and a plurality of wheels mounted to the frame for guided mounting of the heat exchanger panel on the rack structure including first and second upper wheels spaced apart from one another and first and second lower wheels spaced apart from one another. Each upper wheel engages a corresponding upper wheel guiding member of the rack structure, and each lower wheel engages a horizontal portion of a corresponding lower wheel guiding member of the rack structure so that the heat exchanger panel is translatable horizontally. A method for mounting a heat exchanger panel to a rack structure is also provided.

Abstract: A power supply combination comprises a first power source comprising a first power rail delivering power at a first voltage and a second power rail delivering power at a second voltage. The power supply combination also comprises a second power source comprising a third power rail delivering power at the first voltage and a fourth power rail delivering power at the second voltage. A first power combining circuit is electrically connected to the first power rail and to the third power rail. The first power combining circuit delivers power at the first voltage to a first voltage input of a load. A second power combining circuit is electrically connected to the second power rail and to the fourth power rail. The second power combining circuit delivers power at the second voltage to a second voltage input of the load.

Abstract: A heat exchanger assembly has a frame including first and second legs laterally spaced apart, a lower transversal member extending laterally and interconnecting the legs, first and second upstanding members laterally spaced apart and extending upwardly from the lower transversal member, an upper transversal member extending laterally and disposed above the lower transversal member and connected to the first and second upstanding members, and first and second upper retaining members laterally spaced apart and being connected to the upper transversal member and extending transversally to the upper transversal member. First and second heat exchanger panels are connected to the first and second upper retaining members. The heat exchanger panels are disposed in a V-configuration such that a distance between upper ends thereof is greater than a distance between lower ends thereof. A fan has a fan rotation axis extending generally parallel to the first and second upstanding members.

Abstract: A method is devised of deploying a network configuration in a datacenter. The network configuration includes one of more points of interconnection, a point of presence of the datacenter being one of the one or more points of interconnection. Each of the one or more points of interconnection in the network configuration is modeled through objects comprising, for each given point of interconnection, a node object representing the point of interconnection, an interface object hierarchically inferior to the node object and representing a connection to the point of interconnection, an evpnEdge object representing the transport to/from of the point of interconnection in the network, and a layer object hierarchically inferior to the evpnEdge object, and representing the characteristics of the protocol of transport to/from the point of interconnection. The network configuration is set up by a succession of commands on the objects and then pushed to the datacenter.

Abstract: A method and system for detecting anomalies in a data pipeline are disclosed. Data tables are extracted from data pipelines. A best time column is calculated for each data table. Table lines are aggregated so as to establish counts of data lines per a time interval, and corresponding sequences of points are produced that are being analysed through regression methods. An anomaly is raised when points have counts of data lines that lie outside confidence intervals around expected values for such counts.

Abstract: A heat exchanger assembly includes a heat exchanger panel disposed at an inclined orientation. A fan assembly is disposed vertically above the heat exchanger panel and includes a fan impeller connected to a fan mount. The fan impeller is sized and positioned such that part of the fan impeller rotates vertically above the upper end of the heat exchanger panel. A casing has a plurality of inner walls for guiding air from the heat exchanger panel toward the fan assembly. The inner walls include a sloped wall. A distance between an upper end of the sloped wall and a fan rotation axis is greater than a distance between a lower end of the sloped wall and the fan rotation axis. The sloped wall is adjacent to the upper end of the heat exchanger panel such that the part of the fan impeller rotates vertically above the sloped wall.

Abstract: Methods and systems identify connection paths between a power source and a plurality of loads. A reachability signal is sent on a cable connecting a power output of the power source to a power distribution unit (PDU). The PDU receives the reachability and forwards a modified reachability signal on one or more power outputs of the PDU to one or more loads. The PDU receives a return reachability signal from the one or more loads and forwards corresponding modified return reachability signals on the cable to the power source. The power output of the power source receives a plurality of modified return reachability signals that each comprises an identity of the PDU, an identity of a respective power output of the PDU, and an identity of a respective load. A reference between these identities is stored by the power source in a database.