Browse Prior Art Database

performance management method of the large-scale online transaction by getting rid of dependence to quantity of data by adopting 14 variable buckets

IP.com Disclosure Number: IPCOM000225559D
Publication Date: 2013-Feb-18
Document File: 3 page(s) / 86K

Publishing Venue

The IP.com Prior Art Database

Abstract

The interval records which adopted fixed distribution buckets based on average, standard deviation and major variable fixed primary percentile values (85%, 90%, 95%, 98%, 99%), provide online transaction profile. It enables additional analysis and evaluation by utilizing common PC tools such as EXCEL.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 41% of the total text.

Page 01 of 3

performance management method of the large -scale online transaction by getting rid of dependence to quantity of data by adopting 14 variable buckets

The interval records which adopted fixed distribution buckets based on average, standard deviation and major variable fixed primary percentile values (85%, 90%, 95%, 98%, 99%), provide online transaction profile.

It enables additional analysis and evaluation by utilizing common PC tools such as EXCEL.

The interpretation of these series of data, It can be incorporated into the action of system optimization, development to the implementation of the provisioning functions can be more precise.

The problems/issues that is solved by the invention


It solves the following problems as follows and facilitates understanding of changes in transaction characteristics and identify patterns and frequency to occur.


- A client in a financial industry has several million of transactions in a day. In addition, there are several hundred/sec of transactions during peak hours. This number far exceeds the number that can be analyzed and statistical processing capture all of the data such as EXCEL on the PC. Therefore, understanding of the frequency of transactions and the variation in response time, and estimation of the system cost are difficult using common tools.


- In the function of the online transaction, process costs have been increasing for the improvement of the end-user services as follows: processing logic complexity, number of tables in the database to access, number of records to process and so on.


- Understanding the changes and the trends in processing costs and understanding the conditions and the frequencies of large processing load transaction are necessary for capacity planning. However, coupled with the increase in the number of transactions they are becoming difficult.

It captures the continuity of fragmentary information like CT image is possible by taking advantage of the compression of the data by the interval recording that relieves restrictions of amount of data, and the characteristics of the shortening of the time interval of the distribution data. The interpretation of these series of data, it can be incorporated into the action of system optimization. It also help to the implementation of the provisioning functions can be more precise.

Description of the Prior Technology:


These data are managed by using a perspective such as management packages and log analysis programs on a host of tools (Enterprise Server).

As a management point of view, it is intended to set an average value or a certain percentile value whether it achieves.

The transactions are expected that are represented by a single(simple) function transaction like counter operations of the bank. That is, it is assumed to be close to normal distribution of response time characteristics. (High incidence closer to the average, 95% of the transaction between the range of average ± 2 x standard_deviation.)

Accompanied by...