Browse Prior Art Database

Method to relief the Short-On-Storage situation in a transaction processing system Disclosure Number: IPCOM000247156D
Publication Date: 2016-Aug-11
Document File: 5 page(s) / 85K

Publishing Venue

The Prior Art Database


A method to relief the short-on-storage condition in a transaction processing system is disclosed. The method includes defining transactions of whether it can be killed or not, kill threshold, task priority, etc, and then under a persistent short-on-storage situation, the task that meet certain rules is killed to protect the whole system's availability. The rule is set based on how much contiguous storage could be obtained if the task is killed instead of how much storage the task owns.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 5

Method to relief the Short-On-Storage situation in a transaction processing system

Transaction processing(TP) system is widely used nowadays as a really critical system and platform to support critical businessaround the world like banking, airline, insurance and securities etc. This kind of critical business has high demand on huge throughput and very short response time of each transaction. However, sometimes response time is not as fast as expected, even worse the whole system gets stalled which might cause user's business to suffer big loss.

For the various reasons that cause the system to be hang, Short-On-Storage (SOS) is one of the most commonly seen reasons. How to timely get out of the SOS situation has significant meaning to user.

The core idea of the disclosure is to choose certain task or tasks to be killed based on a predefined policy in a bundle project.

The advantage of this disclosure:

- The continuity of storage is a more reasonable indicator comparing with how much storage a task owns when killing tasks.

- Bundle project make the policy easy to update and manage.

- Fragmentation problem is addressed.

- Users' preferences and system's requirement is considered.

- Performance could be guaranteed.

The implementation of the disclosure contains three main steps:

1 Define transactions:

Following attributes can be defined in transaction definition:

- Purge: Purgeable or not

- Purge threshold: XX(size)contiguous storage

- Priority: XX

(more attributes can be defined, if use desires)

2 After SOS condition is met/detected, the system can perform certain existing methods to relief the SOS situation. If the SOS condition persist, the system will choose tasks that meet the following rules to purge:

(1) Purgeable

(2) Certain amount(set by user) of contiguous storage can be obtained if the task is purged.

(3) Low priority (defined by user)

(4) Within all the tasks that meet the first three rules, choose the one that produc...