TW201523230A - Auto node deployment method and server system therefor - Google Patents

Abstract

An auto node deployment method and a server system are provided. The auto node deployment method is used to the server system which includes a plurality of first nodes. The auto node deployment method comprises receiving the node information of first nodes for the node information of the first nodes; identifying, according to the information of the first nodes, the hardware information of the first nodes by installing the operation system to the first nodes; integrating the hardware information of first nodes to produce a plurality of second nodes; receiving the node information of the second nodes for getting the node type of the second nodes; dispatching, according the node type of the second nodes, the second nodes to the server system with installing the software package to the second nodes; and adjusting, according to a dispatching strategy, the numbers of the second nodes which are dispatching in the server system.

Description

Automatic node configuration method and server system

The invention relates to an automatic node configuration method and a server system, in particular to an automatic node configuration method and a server system for automatically configuring nodes according to a server system load.

With the advancement of technology, Cloud Computing has been widely used in three mainstream services, namely Software as a Service (SaaS), Platform as a Service (PaaS) and infrastructure. Infrastructure as a Service (IaaS). Infrastructure-as-a-service is now increasingly applied to services on the network. Infrastructure-as-a-service provides virtual storage, virtual machines, virtual machine clusters, and virtual data centers ( Virtual Data Center) is for users to use. The user provides network services based on the above services. To start the Preboot execution environment server (PXE server), the Dynamic Host Configuration Protocol Server (DHCP Server), and the Network Time Protocol Server (NTP Server). Domain Name System Server (Domain Name System Server, DNS Server) and abnormality monitoring server (Monitor Server).

The server's network service, its load depends on how many users are connected to the server to use the network service. If the number of users increases sharply, the server will often handle the problem due to excessive load, causing the connection to be too slow or down. However, if the number of users is too small, it will often waste the resources of the server. Also, because the number of users changes from time to time, the above-mentioned network services have gradually shifted to the more flexible infrastructure as a service.

Moreover, the resource configuration of the infrastructure as a service is currently configured in a pre-configured manner. For example, for different network services, configure different resources, such as resource storage or computing power. When the number of users increases or decreases, the administrator of the network service must constantly monitor the load status of the network service and manually adjust the amount of resources used by the network service. As a result, it is often impossible to adjust to the situation and waste human resources in a timely manner.

At present, the load of the network service is estimated for a fixed period of time, and the resource adjustment is performed based on the estimated result at regular intervals. However, due to the unpredictability of network service emergencies, most of the network service load changes cannot be accurately predicted. Timing adjustments often result in more waste of resources, which in turn increases the cost of loss. Therefore, there is still room for improvement in the infrastructure-as-a-service resource configuration.

In view of the above problems, the present invention proposes a basis server The system load automatically configures the node's automatic node configuration method and server system to help the server system automatically adjust the number of nodes according to the load to meet the needs of the server system. Avoid the slow or inefficient load of the server system due to the increase of the load, or the idle system cannot be released immediately for other services when the load of the server system is reduced.

The invention provides an automatic node configuration method, which is suitable for a server system, and the server system has a plurality of first nodes. This automatic node configuration method includes the following steps. Receiving node information of the first node to obtain more node information of the first node. According to the node information of the first node, an operating system is installed for each first node to identify the hardware information of the first node. The hardware information of the first node is integrated to generate a plurality of second nodes. Receiving node information of the second node to obtain a node type of the second node. According to the node type of the second node, a software package data is installed for each second node to configure the second node to the server system, so that the server system is started and operates normally. And according to a configuration strategy, adjust the number of second nodes of the server system configuration.

In an embodiment of the invention, the server system operates with a minimum number of second nodes that are operational.

In an embodiment of the invention, the step of adjusting the number of the second nodes of the server system configuration according to the configuration policy includes the following steps. According to the second node, a node attribute list is established. Detects one of the system load values of the server system. And when the system load value is greater than or equal to a system high load bottleneck value, the number of second nodes configured in the server system is increased.

In an embodiment of the invention, the step of detecting a system load value of the server system includes the following steps. When the system load value is less than a system low load bottleneck value, the number of second nodes configured in the server system is reduced.

The present invention provides a server system including a plurality of first nodes, a plurality of second nodes, a database, and a configuration module. The plurality of second nodes are generated according to the hardware information of the first node. The database is used to store node information of the first node, node information of the second node, and node type of the second node. The configuration module is coupled to the first node, the second node, and the database, and receives node information of the first node to obtain information of the first node, and installs an operation for each first node according to node information of the first node. The system, in order to identify the hardware information of the first node, generate a plurality of second nodes according to the hardware information of the first node, receive node information of the second node, to obtain the node type of the second node, and then according to the second node The node type is installed with a software package data for each second node to configure the second node to the server system to enable the server system to start up and operate normally, and adjust the number of second nodes configured by the server system according to a configuration policy.

In an embodiment of the invention, the server system operates with a minimum number of second nodes that are operational.

In an embodiment of the present invention, the foregoing configuration policy includes: establishing a node attribute list according to the second node, and detecting a system load value of the server system, and increasing the configuration when the system load value is greater than or equal to a system high load bottleneck value. The number of second nodes in the server system.

In an embodiment of the invention, when the system load value is less than the system high load bottleneck value, the number of the second nodes configured in the server system is reduced.

According to the automatic node configuration method and the server system disclosed in the present invention, the number of nodes arranged in the server system is determined by the load state of the server system. First, the node information of the first node in the server system is received to obtain the information of the first node. Then, according to the node information of the first node, an operating system is installed for each first node to identify the hardware information of the first node. And integrating the hardware information of the first node to generate a plurality of second nodes. Receiving node information of the second node to obtain a node type of the second node. According to the node type of the second node, a software package data is installed for each second node to configure the second node to the server system, so that the server system is started and operates normally. And according to a configuration strategy, adjust the number of second nodes of the server system configuration. In this way, when the load of the server system is increased or decreased, the resource requirements of the server system can be effectively and immediately increased to increase the stability of the server system.

The above description of the present invention and the following description of the embodiments of the present invention are intended to illustrate and explain the spirit and principles of the invention.

1, 4‧‧‧ server system

10, 40‧‧‧ first node

11, 41‧‧‧ second node

410‧‧‧Operation node

411‧‧‧ Storage node

412‧‧‧ service node

12. 42‧‧‧ database

13, 43‧‧‧ configuration module

430‧‧‧node attribute list

S201~S206, S301~305‧‧‧ Step procedure

1 is a system frame of a server system according to an embodiment of the present invention Composition.

2 is a flow chart of an automatic node configuration method according to an embodiment of the present invention.

3 is a flow chart of an automatic node configuration method according to another embodiment of the present invention.

Figure 4 is a diagram showing an exemplary architecture of a server system in accordance with an embodiment of the present invention.

The detailed features and advantages of the present invention are set forth in the Detailed Description of the Detailed Description of the invention. The objects and advantages associated with the present invention can be readily understood by those skilled in the art. The following examples are intended to describe the present invention in further detail, but are not intended to limit the scope of the invention.

Please refer to FIG. 1 , which is a system architecture diagram of a server system according to an embodiment of the invention. The server system 1 includes a first node 10, a second node 11, a database 12, and a configuration module 13. The server system 1 can be, for example, a Cloud Service Provision Server. The server system 1 is used to provide cloud services such as software as a service, platform as a service, and infrastructure as a service, and configure corresponding resources according to the service to be executed. The first node 10 is, for example, a basic resource node generated by the server system 1 by the server system 1 under the cloud service architecture. This base resource node only Information on hardware configuration does not work when the operating system is not installed. The second node 11 is, for example, a combination of the first node 10, and the second node 11 can serve, for example, as a service node, a storage node, and an arithmetic node. And before the first node 10 has been combined into the second node 11, the first node 10 needs to install the operating system to identify the hardware information of the first node 10, for example.

In this embodiment, the foregoing service node is used to perform Internet services such as Hypertext Transfer Protocol (HTTP) and Database access services. The foregoing storage node is used to execute cloud data access services, such as a file transfer protocol (FTP) and a cloud data access service such as a cloud hard disk. The aforementioned computing nodes are used to execute cloud computing services, such as cloud computing services such as large data sorting and customer management systems. The database 12 is used, for example, to store node information of the first node 10, node information of the second node 11, and node type of the second node 11. The configuration module 13 performs the configuration of the second node 11 according to the database 12, and the configuration module 13 increases or decreases the number of configured second nodes 11 according to the configuration policy. The configuration module 13 can further establish a node attribute table for recording the classification of the second node 11 to speed up the configuration.

For example, when the server system 1 is started, the processing unit of the server system 1 first performs initial setting. After the initialization of the server system 1 is completed, the processing unit of the server system 1 detects the number and information of the first node 10, and the information includes, for example, the network entity address of the baseboard management controller (BMC NIC MAC/IP address). ) and host network entity address (host NIC) MAC/IP address). The foregoing method for detecting the first node 10 is, for example, a processing unit performing a Node Auto Detection Daemon (NAD Daemon). The node automatic detection thread is a thread resident in the server system 1 executed by the processing unit according to a fixed time, and the thread provides an application program interface (API), for example, the application program interface is Interprocess communication (IPC) with the protocol interface (TCP/IP socket) interacts with other threads. After the server system 1 automatically detects the number and information of the first node 10 via the node, the server system 1 stores the quantity and information of the first node 10 in the database 12. The server system 1 can further execute a Node Provisioning Daemon (NP Daemon) via the processing unit, and the node provisioning thread is also a resident thread executed by the processing unit according to a fixed time. The server system 1 further installs the operating system on the first node 10 via the node provisioning thread to identify the hardware information of the first node 10 (for example, the data of the SMBIOS). The installation of the operating system on the first node 10 can be performed, for example, by means of a Preboot eXecution Enviment Boot (PXE boot).

Then, the second node 11 is generated after the first node 10 is integrated by the processing unit via the server system 1, and the second node can be classified into, for example, a service node, a storage node, and an operation node. In the present embodiment, the service node in the second node 11 is composed of, for example, a first node 10. The storage node in the second node 11 consists, for example, of three first nodes 10. Transport in the second node 11 The arithmetic node consists, for example, of two first nodes 10. However, the foregoing number of service nodes, storage nodes and computing nodes is only one embodiment of the present invention, but the present invention is not limited thereto, and the server system 1 can adjust the number of these nodes according to its needs. The node supply thread in the server system 1 also stores the node information of the second node 11 and the node type of the second node 11 in the database 12.

Moreover, the node provisioning thread executed in the server system 1 installs the corresponding software package data for the second node 11 for the node information storing the second node 11 and the node type of the second node 11. The second node 11 is placed in the server system 1 via the configuration module 13, so that the server system 1 starts the service and operates normally. The number of the second nodes 11 configured in the server system 1 is adjusted according to the configuration policy, and the configuration policy is set according to the load of the server system 1. The server system 1 configures a minimum amount of the second node 11 that initiates the service, for example, via a Minimum Cloud Deployment Daemon (MCD Daemon). The server system 1 is, for example, via the On-Demand Cloud Deployment Daemon (DCD Daemon), and the number of second nodes 11 required according to the load configuration.

An automatic node configuration method can be summarized by the description of the above embodiments. Please refer to FIG. 2, which is a flowchart of an automatic node configuration method according to an embodiment of the present invention. The automatic node configuration method of this embodiment can be adapted to a server system. In step S201, the node information of the first node is received to obtain node information of the first node, where the node information includes, for example, a base station management controller network entity address (BMC NIC MAC/IP address) and a master Host NIC MAC/IP address. In step S202, the operating system is installed on each first node according to the node information of the first node to identify the hardware information of the first node, such as a system management basic input/output system (System Management BIOS, SMBIOS) information. In step S203, the hardware information of the first node is integrated to generate a plurality of second nodes. In step S204, the node information of the second node is received to obtain the node type of the second node. In step S205, software package information is installed for each second node according to the node type of the second node, so that the second node is configured to the server system, so that the server system is started and operates normally. In step S206, the number of second nodes of the server system configuration is adjusted according to the configuration policy.

Please refer to FIG. 3, which is a flowchart of an automatic node configuration method according to another embodiment of the present invention. In step S301, a node attribute list is established according to the second node. In step S302, the system load value of the server system is detected. In step S303, it is determined whether the system load value is greater than or equal to the system high load bottleneck value. If the system load value is greater than or equal to the system high load bottleneck value, the process proceeds to step S304. On the other hand, if the system load value is less than the system high load bottleneck value, the process proceeds to step S305. In step S304, the number of second nodes configured in the server system is increased. In step S304, the number of second nodes arranged in the server system is reduced.

Please refer to FIG. 4, which is an exemplary architectural diagram of a server system according to an embodiment of the present invention. Most of the components in the figure are the same as in Figure 1, and the new components will be detailed later. When the server system 4 is booted up After the completion, the first node 40 has been detected by the processing unit and installed the operating system. The first node 40 has been integrated into the type and information, and the second node 41 generates the computing node 410, the storage node 411, and the service node 412. The configuration module 43 also generates a node attribute list 430 for recording the classification of the second node 41, such as the operation node 410, the storage node 411, and the service node 412.

First, the server system 4 is, for example, a web server, and its server system 4 is constantly changed in load. In each time period, the habits and numbers of clients browsing the web are different. The service load upper limit in the configuration policy can be preset to 80%, and the service load lower limit is 20%. The server system 4 is, for example, a resident thread having a timing detection service load, and the resident thread monitors the rise and fall of the service load of the server system 4 at any time. When the server system 4 predicts that the service load is about to be higher than 80%, the configuration module 43 increases the number of service nodes 412 in the server system 4 via the node attribute list 430. When the server system 4 predicts that the service load is about to be less than 20%, the configuration module 43 reduces the number of service nodes 412 in the server system 4 via the node attribute list 430. In this way, the server system 4 can adjust the performance of accommodating the network connection by adjusting the number of service nodes 412. When the web browsing connection of the server system 4 suddenly increases, the server system 4 may not be able to load the performance. When the webpage browsing connection of the server system 4 is not much, the resources can be released for use by other services.

Further, the server system 4 is, for example, a file storage server whose server system 4 continuously receives uploaded files. But the space of the server system 4 Limited, if you want to increase the space, you need more configuration storage node 411. The storage load limit of the configuration policy can be set to 90%, and the storage load lower limit is 10%. The server system 4 is, for example, a resident thread having a timing detection storage load, and the resident thread monitors the storage load rise and fall of the server system 4 at any time. When the server system 4 predicts that the storage load is about 90% higher, the configuration module 43 increases the number of storage nodes 411 in the server system 4 via the node attribute list 430. When the predicted storage load of the server system 4 is about 10% lower, the configuration module 43 reduces the number of storage nodes 411 in the server system 4 via the node attribute list 430. As such, the server system 4 can adjust the space that the server system 4 can provide for use by increasing or decreasing the number of storage nodes 412. When the storage space of the server system 4 is insufficient, the server system 4 is not insufficient in capacity. When the storage space of the server system 4 is idle, the space can be released for use by other files.

Furthermore, the server system 4 is, for example, a movie transfer server, and its server system 4 continuously performs a large number of operations for a large number of movies. However, the central processor of the server system 4 has a limited amount of computation. To increase the performance, more computational nodes 410 are required. The upper limit of the computational load of the configuration policy can be set to 70%, and the lower limit of the computational load is 30%. The server system 4 is, for example, a resident thread having a timing detection operation load, and the resident thread monitors the rise and fall of the calculation load of the server system 4 at any time. When the server system 4 predicts that the computational load is about to be above 70%, the configuration module 43 increases the number of compute nodes 410 in the server system 4 via the node attribute list 430. When the server system 4 When the predicted storage load is about 30%, the configuration module 43 reduces the number of computing nodes 410 in the server system 4 via the node attribute list 430. In this way, the server system 4 can adjust the computing performance that the server system 4 can provide by the number of the increase and decrease operation nodes 410. When the computing performance of the server system 4 is insufficient, the server system 4 may not reach the operational bottleneck. When the computing performance of the server system 4 is not used, the computing performance can be released for use by other programs.

According to the automatic node configuration method and the server system disclosed in the present invention, the number of nodes arranged in the server system is determined by the load state of the server system. The node information of the first node in the server system is first received to obtain the node information of the first node. Then, according to the node information of the first node, an operating system is installed for each first node to identify the hardware information of the first node. And integrating the hardware information of the first node to generate a plurality of second nodes. Receiving node information of the second node to obtain a node type of the second node. According to the node type of the second node, a software package data is installed for each second node to configure the second node to the server system, so that the server system is started and operates normally. And according to a configuration strategy, adjust the number of second nodes of the server system configuration. In this way, when the load of the server system is increased or decreased, the resource requirements of the server system can be effectively and immediately increased to increase the stability of the server system.

While the invention has been described above in the foregoing embodiments, it is not intended to limit the invention. Without departing from the spirit and scope of the invention Both the movement and the retouching are within the scope of patent protection of the present invention. Please refer to the attached patent application for the scope of protection defined by the present invention.

1‧‧‧Server system

10‧‧‧ first node

11‧‧‧second node

12‧‧‧Database

13‧‧‧Configuration module

Claims (8)

An automatic node configuration method is suitable for a server system, the server system has a plurality of first nodes, and the automatic node configuration method includes: receiving node information of the first nodes to obtain nodes of the first nodes Information: according to the node information of the first nodes, installing an operating system for each of the first nodes to identify hardware information of the first nodes; integrating hardware information of the first nodes to generate a plurality of second nodes; receiving node information of the second nodes to obtain node types of the second nodes; and installing a software package data for each of the second nodes according to node types of the second nodes And configuring the second nodes to the server system to enable the server system to start up and operate normally; and adjusting the number of the second nodes configured by the server system according to a configuration policy. The automatic node configuration method of claim 1, wherein the server system operates with a minimum number of the second nodes operable. The automatic node configuration method of claim 1, wherein the step of adjusting the number of the second nodes configured by the server system according to the configuration policy comprises: establishing a node attribute list according to the second nodes; Detecting a system load value of the server system; and increasing the number of the second nodes configured in the server system when the system load value is greater than or equal to a system high load bottleneck value. The automatic node configuration method of claim 3, wherein the step of detecting the system load value of the server system comprises: when the system load value is less than a system low load bottleneck value, reducing configuration on the server system The number of these second nodes. A server system, comprising: a plurality of first nodes; a plurality of second nodes are generated according to the hardware information of the first nodes, and a database is used for storing the first nodes Node information, node information of the second nodes, and node types of the second nodes; and a configuration module, coupled to the first nodes, the second nodes, and the database, and receiving the first Node information of the node to obtain node information of the first nodes, and installing an operating system for each of the first nodes according to the node information of the first nodes to identify the hardware of the first nodes Information, generating a plurality of second nodes according to the hardware information of the first nodes, receiving node information of the second nodes, to obtain node types of the second nodes, and then according to the node types of the second nodes Installing a software package information for each of the second nodes to The second nodes are configured to the server system to enable the server system to start up and operate normally, and the number of the second nodes configured by the server system is adjusted according to a configuration policy. The server system of claim 5, wherein the server system operates with a minimum number of the second nodes operable. The server system of claim 5, wherein the configuration policy comprises: establishing a node attribute list according to the second nodes, and detecting the system load value of the server system, when the system load value is greater than or equal to one When the system has a high load bottleneck value, the number of the second nodes configured in the server system is increased. The server system of claim 5, wherein when the system load value is less than the system high load bottleneck value, the number of the second nodes configured in the server system is reduced.