Shenzhen Microbt Electronics Technology Co., Ltd.

Lihong Huang

Zuoxing Yang

Weibin Ma

Haifeng Guo

Yuefeng Wu

Jianbo Liu

Hong Zhang

Zhiming Fu

Jianli Wei

This disclosure relates to a method, an apparatus, and a data processing device for controlling a frequency search speed, the method for controlling the frequency search speed including: obtaining parameter information of a frequency search speed regulation gear; obtaining a frequency search parameter based on the parameter information of the frequency search speed regulation gear; and performing frequency search based on the obtained frequency search parameter.

Method, apparatus, data processing device, and medium for controlling frequency search speed

Embodiments of this application provide a chip frequency control method and apparatus, a blockchain server, and a storage medium. The method includes: determining a real-time computing power ratio of a chip of a blockchain server, a real-time temperature of the chip, and a real-time voltage of the chip; and adjusting a setting frequency of the chip based on a first comparison result of the real-time computing power ratio with a reference computing power ratio of the chip, a second comparison result of the real-time temperature with a reference temperature of the chip, and a third comparison result of the real-time voltage with a reference voltage of the chip, wherein the reference computing power ratio, the reference temperature, and the reference voltage are determined when the blockchain server undergoes a frequency rising phase after start-up and enters a working status. When an exception occurs in the chip, timely and appropriate response can improve stability of the blockchain server.

Chip frequency control method and apparatus, blockchain server, and storage medium

Disclosed is a method for determining configuration parameters of a data processing device, including: operating the data processing device by using configuration parameters, which are universal optimization configuration parameters obtained according to a universal operating parameter model; during the operating process, changing the configuration parameters to obtain a dedicated operating parameter data set, which includes a plurality of groups of operating parameters, each of which includes configuration parameters and capability parameters of the data processing device when the data processing device is operating under the configuration parameters; executing model training by using the dedicated operating parameter data set to obtain a dedicated operating parameter model; and obtaining optimal configuration parameters according to the dedicated operating parameter model, and operating the data processing device according to the optimal configuration parameters, where the optimal configuration parameters are configuration parameters corresponding to optimal capability parameters in the dedicated operating parameter data set.

Method for determining configuration parameters of data processing device, electronic device and storage medium

The present disclosure relates to a method for providing clock frequencies for computing cores, a chip and a data processing device. The method includes: causing a main clock frequency unit to provide a first main clock frequency for computing cores; testing the computing cores operating at the first main clock frequency to determine whether a pass rate of the computing cores is greater than an upper threshold or less than a lower threshold; when the pass rate is less than the lower threshold, causing an auxiliary clock frequency unit to provide a lower first auxiliary clock frequency for computing cores abnormally operating, causing the main clock frequency unit to providing the first main clock frequency for the remaining computing cores; when the pass rate is greater than the upper threshold, causing the main clock frequency unit to provide a higher second main clock frequency for the computing cores.

Method for providing clock frequencies for computing cores, chip and data processing device

Implementations of this application provide a method and an apparatus for controlling a voltage of a power supply of a data processing device and a data processing device. The method includes: determining a computing power ratio of the data processing device based on an actual computing power and a theoretical computing power of the data processing device; generating a power supply control instruction based on a result of comparison between the computing power ratio and a predetermined threshold; and controlling an output voltage of the power supply of the data processing device based on the power supply control instruction. According to the implementations of this application, the output voltage of the power supply is controlled according to the computing power ratio, and a good compromise can be obtained between the power consumption loss and the computing power of the data processing device.

Method and apparatus for controlling voltage of power supply of data processing device, data processing device, and storage medium

A method and apparatus for starting up a digital currency data processing device, and a digital currency data processing device. The digital currency data processing device includes a hash board including a plurality of hash chip groups. The method includes: transmitting a startup signal to a refrigerating part of a data processing device before turning on a power supply; turning on the power supply; and controlling respective hash chips in each hash chip group to gradually and synchronously turn on cores. The above arrangement improves voltage balance of a hash board, and also ensures the accuracy of a clock signal.

Method and apparatus for starting up digital currency data processing device, and digital currency data processing device

The present disclosure relates to a method for up-converting a clock signal, a clock circuit and a digital processing device. More specifically, provided is a method for up-converting a clock signal, comprising: employing a first clock sub-circuit to provide a clock signal having a first frequency to a chip; receiving an instruction to up-convert the clock signal having the first frequency to a clock signal having a second frequency; in response to receiving the instruction, causing a second clock sub-circuit to output the clock signal having the second frequency; and after the second clock sub-circuit outputs the clock signal having the second frequency, employing the second clock sub-circuit to provide the clock signal having the second frequency to the chip in place of the first clock sub-circuit.

Method for up-converting clock signal, clock circuit and digital processing device

Implementations of the present application propose a method and apparatus for preventing rollback of firmware of a data processing device, and a data processing device. The method includes: enabling a boot loader (BootLoader) to read a current value of a predetermined bit in a one-time programmable memory (eFuse); determining whether the current value and a legal value written into the one-time programmable memory after the latest updating of the firmware of the data processing device satisfy a preset relationship; in response to determining that the current value and the legal value satisfy the preset relationship, enabling the boot loader to call an operating system kernel of the data processing device, and in response to determining that the current value and the legal value do not satisfy the preset relationship, enabling the boot loader not to call the operating system kernel of the data processing device. According to the implementations of the present application, rollback of the firmware can be prevented based on a variety of ways in a link-by-link mode.

Method and apparatus for preventing rollback of firmware of data processing device, and data processing device

The present disclosure discloses a rapid frequency searching method and apparatus for a data processing device, and a data processing device. The method includes: gradually raising a frequency of a computing chip from an initial frequency point to a target frequency through at least one first frequency point at an initial output voltage of a power source of the data processing device, wherein the duration when the computing chip is located at each of the at least one first frequency point is shorter than the duration when the computing chip is located at each of the at least one first frequency point in a routine frequency searching process; and after the computing chip achieves the target frequency, adjusting the output voltage of the power source of the data processing device and the frequency of the computing chip according to at least one output voltage value and at least one second frequency point in combination with test tasks delivered to the computing chip and feedback results from the computing chip to obtain a limit frequency of the computing chip and an output voltage, corresponding to the limit frequency, of the power source of the data processing device.

Rapid frequency searching method and apparatus for data processing device, and data processing device

Disclosed is a mining machine power adjusting method, including: when a mining machine is in a working stage, collecting an input voltage of the mining machine; adjusting a maximum working power of the mining machine to a first power value by controlling a working frequency and an output voltage of the mining machine when the input voltage is lower than a pre-determined voltage threshold; and adjusting the maximum working power to a second power value by controlling the working frequency and the output voltage of the mining machine when the input voltage is higher than or equal to the voltage threshold, where the first power value is the product of the input voltage multiplied by a rated current of the mining machine, and the second power value is the product of the voltage threshold multiplied by the rated current. According to the disclosure, the power of the mining machine is adjusted according to changes in the input voltage by fully utilizing a voltage range slightly higher than a standard input voltage, so that the calculation power of the mining machine can be greatly increased to achieve better mining revenues without significantly increasing the power consumption ratio of the mining machine while ensuring the stable and safe running of the mining machine.

Growing community of inventors

Shenzhen, China

Lihong Huang