Multi-multidimensional computer architecture for big data applications

Patent metadata

Field	Value
Publication number	US-10417005-B2
Application number	US-201715700488-A
Country	US
Kind code	B2
Filing date	Sep 11, 2017
Priority date	Mar 17, 2015
Publication date	Sep 17, 2019
Grant date	Sep 17, 2019

Abstract

A data processing apparatus is provided comprising a front-end interface electronically coupled to a main processor. The front-end interface is configured to receive data stored in a repository, in particular an external storage and/or a network, determine whether the data is a single-access data or a multiple-access data by analyzing an access parameter designating the data, route the multiple-access data for processing by the main processor, and route the single-access data for pre-processing by the front-end interface and routing results of the pre-processing to the main processor.

First claim

The invention claimed is:

1. A data processing apparatus, comprising:
a front-end interface electronically coupled to a main processor; the front-end interface is configured to:
receive input data from a data repository, wherein the data repository includes one of an external storage device and/or a network repository accessible using a network interface;
determine whether the input data is a single-access data or a multiple-access data by analyzing an access parameter designating the input data;
route the multiple-access data for processing by the main processor; and
route the single-access data for pre-processing by the front-end interface and routing results of the pre-processing to the main processor.

2. The data processing apparatus of claim 1 , wherein the access parameter is based on a reception time between two receptions of a same data on the front-end interface, or a request time between two requests of a same data by the main processor.

3. The data processing apparatus of claim 2 , wherein the access parameter is measured as a relative time between two receptions and/or two request of a same data on the front-end interface.

4. The data processing apparatus of claim 2 , wherein the access parameter is measured as an absolute time of a second access of a same data on the front-end interface.

5. The data processing apparatus of claim 2 , wherein the access parameter is measured by counting a number of data accesses between two accesses of a same data on the front-end interface.

6. The data processing apparatus of claim 2 , wherein the single access data is determined by comparing the access parameter to a threshold value.

7. The data processing apparatus of claim 1 , wherein the front-end interface comprises at least one local processor adapted to:
access a local memory unit;
pre-process the single-access data and/or store the single-access data to the local memory unit; and
route results of the pre-processing to the main processor for further processing.

8. The data processing apparatus of claim 1 , wherein the pre-processing comprises a group of application specific instructions for filtering and aggregating a plurality of data elements of the single-access data.

9. The data processing apparatus of claim 8 , wherein the at least one local processor of the front-end interface incorporates a pre-processing instruction transfer module, wherein the pre-processing instruction transfer module is configured to receive the group of application specific instructions from the main processor, convert the group of application specific instructions to a machine language of the at least one processor, and execute the machine language on the at least one processor to perform the pre-processing.

10. The data processing apparatus of claim 1 , wherein the front-end interface comprises a temporary memory unit for storing at least some members of the results before routed to the main processor.

11. The data processing apparatus of claim 10 , wherein memory storage addressed in the temporary memory unit is mapped into storage addresses of a main memory of the main processor.

12. The data processing apparatus of claim 1 , wherein the front-end interface is adapted to receive a main processor output from the main processor and to use the main processor output to pre-process the single-access data.

13. The data processing apparatus of claim 1 , wherein a local processor instruction is generated by a device driver incorporated into an operating system configured to operate the main processor to perform the pre-processing.

14. A method, comprising:
receiving input data stored in a repository, wherein the data repository includes one of an external storage device and/or a network repository accessible using a network interface;
determining whether the input data is a single-access data or a multiple-access data by

analyzing an access parameter designating the data;
routing the multiple-access data for processing by a main processor; and
routing the single-access data for pre-processing by a front-end interface and routing results of the pre-processing to the main processor.

15. A non-transitory computer readable media storing computer instructions, that when executed by one or more processors, cause the one or more processors to perform the steps of:
receive input data stored in a repository, wherein the data repository includes one of an external storage device and/or a network repository accessible using a network interface;
determine whether the input data is a single-access data or a multiple-access data by analyzing an access parameter designating the data;
route the multiple-access data for processing by a main processor; and
route the single-access data for pre-processing by a front-end interface and routing results of the pre-processing to the main processor.

Assignees

• Huawei Tech Co Ltd

Inventors

• Weiser Uri
• Horowitz Tal
• Wang Jintang

Classifications

• G06F9/5027 Primary

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06F9/3895 Primary

  for complex operations, e.g. multidimensional or interleaved address generators, macros · CPC title

• G06F9/5083

  Techniques for rebalancing the load in a distributed system · CPC title

• G06F15/17

  using an input/output type connection, e.g. channel, I/O port · CPC title

• G06F9/5055

  considering software capabilities, i.e. software resources associated or available to the machine · CPC title