# US 2016/0259812 A1 · Method and system for accessing a distributed file system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2016/0259812 A1
- Document ID: US20160259812A1
- Status: Published application
- Kind: A1
- Application number: 15/157,188
- Filed: 2016-05-17
- Published: 2016-09-08
- Earliest claimed: 2013-12-11
- Assignees: Tencent Technology (Shenzhen) Company Limited
- Applicants: Tencent Technology (Shenzhen) Company Limited
- Inventors: Yunjia WU
- Web page: https://patentgrep.com/patent/US20160259812A1

## CPC

- G06F 16/183: {Provision of network file services by network file servers, e.g. by using NFS, CIFS (network file access protocols H04L67/1097)}
- H04L 67/06: specially adapted for file transfer, e.g. file transfer protocol [FTP]
- H04L 67/1027: Persistence of sessions during load balancing
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A method and system for accessing an HDFS is disclosed. In some embodiments, a method of providing access to an HDFS includes: at an interface server having one or more processors and memory: maintaining a respective connection pool comprising a respective plurality of persistent connections to the HDFS; receiving respective access requests from a plurality of accessing devices to access the HDFS; creating a corresponding Hadoop operation for each of the respective access requests received from the plurality of accessing devices; and executing the corresponding Hadoop operation created for each of the respective access requests through a respective set of persistent connections selected from the respective connection pool maintained by the interface server.

## Claims

### Claim 1

A method of providing access to a Hadoop Distributed File System (HDFS), comprising:
at an interface server having one or more processors and memory:
maintaining a respective connection pool comprising a respective plurality of persistent connections to the HDFS;
receiving respective access requests from a plurality of accessing devices to access the HDFS;
creating a corresponding Hadoop operation for each of the respective access requests received from the plurality of accessing devices; and
executing the corresponding Hadoop operation created for each of the respective access requests through a respective set of persistent connections selected from the respective connection pool maintained by the interface server.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
for a first access request of the received respective access requests:
identifying a respective set of data nodes of the HDFS that are targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include persistent connections between the interface server and at least some of the respective set of data nodes.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
for a first access request of the received respective access requests:
identifying a respective namenode of the HDFS that is targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include a persistent connection between the interface server and the respective namenode.

### Claim 4 (depends on 1)

The method of claim 1, wherein the interface server includes a respective Hadoop client for the HDFS, and the corresponding Hadoop operation is executed by the Hadoop client.

### Claim 5 (depends on 1)

The method of claim 1, wherein the corresponding Hadoop operation is executed by one or more command line inputs provided from the interface server to the HDFS.

### Claim 6 (depends on 1)

The method of claim 1, wherein the corresponding Hadoop operation is executed by a corresponding program provided from the interface server to the HDFS.

### Claim 7 (depends on 1)

The method of claim 1, wherein the interface server is one of a plurality of interface servers connected to the HDFS, and the plurality of interface servers each maintains a respective connection pool comprising a respective plurality of persistent connections to the HDFS.

### Claim 8 (depends on 7)

The method of claim 7, wherein the respective access requests from the plurality of accessing devices are transmitted to the interface server based on a predetermined load balancing criterion established for the plurality of interface servers.

### Claim 9

An interface server for providing access to a Hadoop Distributed File System (HDFS), comprising:
one or more processors; and
memory having instructions stored thereon, the instructions, when executed by the one or more processors, cause the processors to perform operations comprising:
maintaining a respective connection pool comprising a respective plurality of persistent connections to the HDFS;
receiving respective access requests from a plurality of accessing devices to access the HDFS;
creating a corresponding Hadoop operation for each of the respective access requests received from the plurality of accessing devices; and
executing the corresponding Hadoop operation created for each of the respective access requests through a respective set of persistent connections selected from the respective connection pool maintained by the interface server.

### Claim 10 (depends on 9)

The interface server of claim 9, wherein the operations further comprise:
for a first access request of the received respective access requests:
identifying a respective set of data nodes of the HDFS that are targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include persistent connections between the interface server and at least some of the respective set of data nodes.

### Claim 11 (depends on 9)

The interface server of claim 9, wherein the operations further comprise:
for a first access request of the received respective access requests:
identifying a respective namenode of the HDFS that is targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include a persistent connection between the interface server and the respective namenode.

### Claim 12 (depends on 9)

The interface server of claim 9, wherein the interface server includes a respective Hadoop client for the HDFS, and wherein executing the corresponding Hadoop operation further comprises executing the corresponding Hadoop operation using the Hadoop client.

### Claim 13 (depends on 9)

The interface server of claim 9, wherein executing the corresponding Hadoop operation further comprises:
executing the corresponding Hadoop operation by providing one or more command line inputs to the HDFS.

### Claim 14 (depends on 9)

The interface server of claim 9, wherein executing the corresponding Hadoop operation further comprises:
providing a respective program for the corresponding Hadoop operation to the HDFS.

### Claim 15 (depends on 9)

The interface server of claim 9, wherein the interface server is one of a plurality of interface servers connected to the HDFS, and the plurality of interface servers each maintains a respective connection pool comprising a respective plurality of persistent connections to the HDFS.

### Claim 16 (depends on 15)

The interface server of claim 15, wherein the respective access requests from the plurality of accessing devices are transmitted to the interface server based on a predetermined load balancing criterion established for the plurality of interface servers.

### Claim 17

A non-transitory computer-readable medium having instructions stored thereon, the instructions, when executed by one or more processors, cause the processors to perform operations comprising:
at an access server that provides access to a Hadoop Distributed File System (HDFS):
maintaining a respective connection pool comprising a respective plurality of persistent connections to the HDFS;
receiving respective access requests from a plurality of accessing devices to access the HDFS;
creating a corresponding Hadoop operation for each of the respective access requests received from the plurality of accessing devices; and
executing the corresponding Hadoop operation created for each of the respective access requests through a respective set of persistent connections selected from the respective connection pool maintained by the access server.

### Claim 18 (depends on 17)

The computer-readable medium of claim 17, wherein the operations further comprise:
for a first access request of the received respective access requests:
identifying a respective set of data nodes of the HDFS that are targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include persistent connections between the interface server and at least some of the respective set of data nodes.

### Claim 19 (depends on 17)

The computer-readable medium of claim 17, wherein the operations further comprise:
for a first access request of the received respective access requests:
identifying a respective namenode of the HDFS that is targeted by the corresponding Hadoop operation generated for the first access request, wherein the respective set of persistent connections include a persistent connection between the interface server and the respective namenode.

### Claim 20 (depends on 17)

The computer-readable medium of claim 17, wherein the interface server includes a respective Hadoop client for the HDFS, and wherein executing the corresponding Hadoop operation further comprises executing the corresponding Hadoop operation using the Hadoop client.
