Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

32 results about "Scalable design" patented technology

“Scalable design is a web design solution, it is an turning point that urges designers and developers to become involved, make meaningful contributions and embrace the simplicity for a more productive future”.

Cycle segmented prefix circuits

The poor scalability of existing superscalar processors has been of great concern to the computer engineering community. In particular, the critical-path delays of many components in existing implementations grow quadratically with the issue width and the window size. This patent presents a novel way to reimplement these components and reduce their critical-path delay growth. It then describes an entire processor microarchitecture, called the Ultrascalar processor, that has better critical-path delay growth than existing superscalars. Most of our scalable designs are based on a single circuit, a cyclic segmented parallel prefix (cspp). We observe that processor components typically operate on a wrap-around sequence of instructions, computing some associative property of that sequence. For example, to assign an ALU to the oldest requesting instruction, each instruction in the instruction sequence must be told whether any preceding instructions are requesting an ALU. Similarly, to read an argument register, an instruction must somehow communicate with the most recent preceding instruction that wrote that register. A cspp circuit can implement such functions by computing for each instruction within a wrap-around instruction sequence the accumulative result of applying some associative operator to all the preceding instructions. A cspp circuit has a critical path gate delay logarithmic in the length of the instruction sequence. Depending on its associative operation and its layout, a cspp circuit can have a critical path wire delay sublinear in the length of the instruction sequence.
Owner:YALE UNIV

Plug-in mechanism-based virtual machine migration method

The invention discloses a plug-in mechanism-based virtual machine migration method. According to the plug-in mechanism-based virtual machine migration method, a source virtual machine management platform and a target virtual machine management platform are provided, and a migration agent module, a driving module and an import guide module are designed. Based on the three modules, the particular migration process of the virtual machine is as follows: the migration agent module acquires a virtual machine disk image and a configuration specification parameter of the virtual machine from the source virtual machine management platform and finishes conversion; the driving module is used for managing and driving a plug-in; the import agent module rebuilds a virtual machine on the target virtual machine management platform by using the converted virtual machine disk image. Compared with the prior art, the plug-in mechanism-based virtual machine migration method has the advantages that automatic migration of the cross-platform virtual machine is realized; the virtual machine migration efficiency in the operating and maintaining process of the virtual machine management platform is effectively improved; the method is based on an extensible design architecture, so that the new virtual machine management platform can be supported without modifying the original system, and the practicability is high.
Owner:SHANDONG INSPUR CLOUD SERVICE INFORMATION TECH CO LTD

Ultra-long distance distribution type optical fiber sensor and using method thereof

The invention discloses an ultra-long distance distribution type optical fiber sensor and a method for measuring temperature and vibration signals. The ultra-long distance distribution type optical fiber sensor adopts a multi-stage relay amplifier, and a high-power pulse light source enters a detecting optical fiber stage by stage through proportional allocation of a plurality of optical fiber couplers and transmission of transmission fibers, and re-amplifies the pulse of the detecting light at a specified length, thereby improving the signal to noise ratio of the system in terms of 'source', being capable of avoiding non-linear scattering caused by overrun light which enters a detecting optical cable directly on one hand, and being capable of reducing higher energy loss caused by continuous light transmission on the other hand. The ultra-long distance distribution type optical fiber sensor adopts the staged design, thereby greatly reducing the requirement on receiving sensitive and dynamic range of a receiving circuit. In addition, the ultra-long distance distribution type optical fiber sensor adopts the expandable design, thereby being capable of conveniently realizing short-distance detection, long-distance detection and ultra-long distance detection in sequence.
Owner:SHANGHAI BOOM FIBER SENSING TECH

Efficient circuits for out-of-order microprocessors

The poor scalability of existing superscalar processors has been of great concern to the computer engineering community. In particular, the critical-path delays of many components in existing implementations grow quadratically with the issue width and the window size. This patent presents a novel way to reimplement these components and reduce their critical-path delay growth. It then describes an entire processor microarchitecture, called the Ultrascalar processor, that has better critical-path delay growth than existing superscalars. Most of our scalable designs are based on a single circuit, a cyclic segmented parallel prefix (cspp). We observe that processor components typically operate on a wrap-around sequence of instructions, computing some associative property of that sequence. For example, to assign an ALU to the oldest requesting instruction, each instruction in the instruction sequence must be told whether any preceding instructions are requesting an ALU. Similarly, to read an argument register, an instruction must somehow communicate with the most recent preceding instruction that wrote that register. A cspp circuit can implement such functions by computing for each instruction within a wrap-around instruction sequence the accumulative result of applying some associative operator to all the preceding instructions. A cspp circuit has a critical path gate delay logarithmic in the length of the instruction sequence. Depending on its associative operation and its layout, a cspp circuit can have a critical path wire delay sublinear in the length of the instruction sequence.
Owner:YALE UNIV

Method and device for controlling high-speed serial storage

The invention discloses a method and device for controlling high-speed serial storage. The method includes the steps of S1, defining a data package writing format and/or a data package reading format; S2, building a communication chain of a plurality of data channels, and aligning the multiple data channels; S3, analyzing a data package according to the defined data package writing format and/or the defined data package reading format; S4, carrying out writing/reading on a memory. The device comprises at least one high-speed serial channel, a high-speed serial interface control module, a protocol processing module and a storage control module. According to the method and device, due to protocol conversion and storage control, the perfect combination of the high throughput rate of a high-speed serial interface and rapid reading or rapid writing of a static random access memory is achieved, the extensible design of the number of the channels of the high-speed serial interface and the capacity of the static random access memory is achieved, then a storage body large in capacity, high in speed and high in intensity is built, and the method and device further have the advantages of being simple in interface and high in data reliability. The method and device can be widely applied to various high-speed serial storage control systems.
Owner:SHENZHEN STATE MICROELECTRONICS CO LTD

Highly-decoupled method capable of dynamically managing crawlers

The invention discloses a highly-decoupled method capable of dynamically managing a crawler, which comprises the following steps that: the crawler is divided into two stages of data analysis and new target generation, rules of the two stages corresponding to a collection target are compiled into json data according to a protocol, and the json data are stored into a host end; the host runs a task and sends the task to a client with sufficient resources through a message queue module according to a resource scheduling algorithm, the client receives task information, converts the task information into executable information through the crawler protocol core, runs the executable information through a crawler running module, and finally obtains data; the host end obtains data and a new task, and stores and updates a task pool; and the host end is separated from a crawler server, so that the coupling of the system can be reduced. Therefore, after the functions of the crawler are separated, the complexity of the crawler server can be reduced, and the host end can be modified while the distributed crawler system runs so as to achieve the purpose of specific control management, so that the whole module is subjected to decoupling and extensible design, and the robustness and stability of the whole framework are enhanced.
Owner:苏州市中地行信息技术有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products