Efficient replacement of clients running large scale applications

Patent metadata

Field	Value
Publication number	US-10855521-B2
Application number	US-201514876847-A
Country	US
Kind code	B2
Filing date	Oct 7, 2015
Priority date	Jun 29, 2015
Publication date	Dec 1, 2020
Grant date	Dec 1, 2020

Abstract

A system and method for efficiently replacing clients running a large scale application uses a pool of clients, which includes a group of active clients running the large scale application and at least one standby client. When a particular active client is identified to be replaced, one of the at least one standby client is selected as a replacement client for the particular active client. The particular active client is replaced with the selected replacement client with respect to the running of the large scale application such that the particular client is removed from the group of active clients and the selected replacement client is added to the group of active clients.

First claim

What is claimed is:

1. A method for efficiently replacing clients running a large scale application, the method comprising:
deploying a plurality of clients as a pool of clients to run the large scale application, the pool of clients including a group of active clients that will be running the large scale application and at least one standby client that is not currently active in running the large scale application, a standby client being a client that has been powered down to a standby mode;
identifying a particular client to be replaced from the group of active clients in the pool of clients that are currently running the large scale application; wherein the identifying the particular client to be replaced includes detecting a poor-performing active client from the group of active clients in the pool of clients that are currently running the large scale application based on a performance of one or more operations being executed by the poor-performing active client and a response time of the poor-performing active client for a task request, wherein the poor-performing active client is the particular client to be replaced;
selecting one of the at least one standby client in the pool of clients as a replacement client;
sending a command to wake up the selected replacement client;
replacing the particular client with the selected replacement client with respect to the running of the large scale application such that the particular client is removed from the group of active clients and the selected replacement client is added to the group of active clients;
restarting the poor-performing active client and placing the restarted poor performing active client in the standby mode in the pool of clients when a severity level of the poor-performing active client with respect to performance of the large scale application does not exceed a threshold; and
deleting the poor-performing active client from the pool of clients and initiating a client provisioning request to create a new client in the pool of clients in the standby mode for the large scale application when the severity level of the poor performing active client with respect to performance of the large scale application does exceed the threshold.

2. The method of claim 1 , wherein the poor-performing active client is an active client that is performing below a performance threshold.

3. The method of claim 1 , further comprising:
executing a software upgrade on the particular client after the particular client has been replaced; and
adding the upgraded particular client to the pool of clients as a standby client.

4. The method of claim 1 , wherein the deploying the plurality of clients includes deploying the plurality of clients using a deployment profile such that each of the plurality of clients is deployed using the same deployment profile.

5. The method of claim 1 , wherein the replacing the particular client with the selected replacement client includes:
initiating a scale-out operation to add the selected replacement client to the group of active clients; and
initiating a scale-in operation to remove the particular client from the group of active clients.

6. A system for efficiently replacing clients running a large scale application comprising:
memory;
at least one processor programmed to:
deploy a plurality of clients as a pool of clients to run the large scale application, the pool of clients including a group of active clients that will be running the large scale application and at least one standby client that is not currently active in running the large scale application, a standby client being a client that has been powered down to a standby mode;
identify a particular client to be replaced from the group of active clients in the pool of clients that are currently running the large scale application, wherein the identifying the particular client to be replaced includes detecting a poor-performing active client from the group of active clients in the pool of clients that are currently running the large scale application based on a performance of one or more operations being executed by the poor-performing active client and a response time of the poor-performing active client for a task request, wherein the poor-performing active client is the particular client to be replaced;
select one of the at least one standby client in the pool of clients as a replacement client;
send a command to wake up the selected replacement client;
replace the particular client with the selected replacement client with respect to the running of the large scale application such that the particular client is removed from the group of active clients and the selected replacement client is added to the group of active clients;
restart the poor-performing active client and place the restarted poor-performing active client in the standby mode in the pool of clients when a severity level of the poor-performing active client with respect to performance of the large scale application does not exceed a threshold; and
delete the poor-performing active client from the pool of clients and initiate a client provisioning request to create a new client in the pool of clients in the standby mode for the large scale application when the severity level of the poor performing active client with respect to performance of the large scale application does exceed the threshold.

7. The system of claim 6 , wherein the poor-performing active client is an active client that is performing below a performance threshold.

8. The system of claim 6 , wherein the at least one processor is further configured to initiate a scale-out operation to add the selected replacement client to the group of active clients, and to initiate a scale-in operation to remove the particular client from the group of active clients.

9. A non-transitory computer-readable storage medium containing program instructions for a method for efficiently replacing clients running a large scale application, wherein execution of the program instructions by one or more processors of a computer system causes the one or more processors to perform steps comprising:
deploying a plurality of clients as a pool of clients to run the large scale application, the pool of clients including a group of active clients that will be running the large scale application and at least one standby client that is not currently active in running the large scale application, a standby client being a client that has been powered down to a standby mode;
identifying a particular client to be replaced from the group of active clients in the pool of clients that are currently running the large scale application, wherein the identifying the particular client to be replaced includes detecting a poor-performing active client from the group of active clients in the pool of clients that are currently running the large scale application based on a performance of one or more operations being executed by the poor-performing active client and a response time of the poor-performing active client for a task request, wherein the poor-performing active client is the particular client to be replaced;
selecting one of the at least one standby client in the pool of clients as a replacement client;
sending a command to wake up the selected replacement client;
replacing the particular client with the selected replacement client with respect to the running of the large scale application such that the particular client is removed from the group of active clients and the selected replacement client is added to the group of active clients;
restarting the poor-performing active client and placing the restarted poor performing active client in the standby mode in the pool of clients when a severity le

Assignees

• Vmware Inc

Inventors

• Mankala Shyam Sundar Rao
• Singh Servesh
• Holler Anne
• Narayanaswamy Arun
• Singh Kiran

Classifications

• H04L41/0661 Primary

  by reconfiguring faulty entities · CPC title

• H04L41/0668

  by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure · CPC title

• H04L41/0663

  Performing the actions predefined by failover planning, e.g. switching to standby network elements · CPC title

• H04L41/0672 Primary

  Electricity · mapped topic