Control of Performance and Conformance for Data Storage
Introduction
This article has been written to help impliment efficient data storage practices within a medium to large enterprise. As part of any important strategy, the intentions must not be merely stated and then forgotten. Some form of on-going performance and conformance measures must be implemented.
Being a technical project, most of the measures will be of a technical nature. However, it is important to understand the options available and to determine what is the most relevant method for the task.
Measures of Performance
Benchmarking is usually related to technical performance measure. However, it can also be used to measure performance of many disparate endeavours, such as between business competitors. With benchmarking, we must find the best example of a given undertaking and compare our efforts against it. As F. John Reh (2006) states, “Benchmarking is the process of determining who is the very best, who sets the standard, and what that standard is.” As far as our project is concerned, we need to determine the most efficient storage structure and practices for the Public data drive. This has been predetermined as part of the overall project; the structure of the data, the efficient data formats and what data should and should not be stored. The benchmark has already been determined and outlined.
Balanced Scorecard was developed in the early 1990’s by Dr Robert Kaplan and David Norton. They developed it as a “Management System” not merely as a measurement system. As Arveson, P (1998) states, it “enables organizations to clarify their vision and strategy and translate them into action. It provides feedback around both the internal business processes and external outcomes in order to continuously improve strategic performance and results.” His article goes on to explain that we must view the organization from four perspectives: Financial, Internal Business Processes, Learning and Growth, Customer. Each of these viewpoints relates back to the central Vision and Strategy of the organization. The viewpoints provide critical, factual data to the organization in order to make informed, long term decisions.
I feel that in this instance, this approach is somewhat overly complex for the project at hand. However, I understand the strategic importance of the balanced scorecard in a more holistic organizational context.
Results Based Performance Measure is, as Zigon Performance Group states, a “unique way of looking at and teaching others to create measures for the really hard-to-measure aspects of today's work. RBM doesn't ignore behavior or activity, but sees it as the means to the important end - RESULTS.” Therefore, we concentrate on the results, not the measures and procedures implemented to address a shortcoming. Three relevant examples given in the same Zigon article are:
- A car that's repaired and running, not what parts were replaced
- An employee who can do something he couldn't do before, not a training session completed
- A customer who's happy with their purchase, not a checkout procedure that was followed
Conformance issues to consider
Customer/user satisfaction: Ensuring the customer (or user, in this case) is satisfied is paramount to willing conformance. A carefully constructed satisfaction survey will help to measure the relevant satisfaction of customers/users. This can offer tangible benefits in making informed decisions for future evolution of the policies and strategies of the organisation.
Customer/user understanding: If the user does not understand the policies, they will not be able to conform, let alone attempt to conform. Clear policies and instructions are necessary for that understanding. Again, an appropriately written survey of users will identify any problems with understanding.
Level of contingency enforcement: Reduced levels of policy contingency enforcement is a clear indication that the user is conforming to the requirements. Keeping records of contingency enforcement will show trends in conformance. Increased need for enforcement indicates a downturn in conformance with the opposite meaning an increase in conformance.
Measures
As stated above, the most appropriate measure for this project is Benchmarking. The project has already determined the data formats, structure and accountability metrics in the Policy Statement, thus we can monitor compliance using a number of technical methods.
The Removal of Stale and/or Irrelevant Data
Measure of performance:
- No data older than six months.
- No duplicate files.
- No irrelevant or inappropriate data.
Measure of conformance:
- Less than five documents per month older than six months.
- Less than two sets of duplicate files per month.
- Less than two files deemed irrelevant per month.
Method of measure:
A global search of the Public drive in date order will be used to determine the age of data. Any files showing as older than six months will be actioned. Any files less than a month old will be reviewed for relevance. Duplicate finding software will be run on the Public drive and any duplicates flagged for action.
The Education of Users as to Appropriate Storage Practices
Measure of performance:
- No BMP, TIF, MOV, WMV, WMA, MP3 files on the drive.
- No JPG files over 500Kb and no Documents over 1Mb.
- All low-value (persistent but rarely accessed) data is compressed or removed.
Measures of conformance:
- Less than five BMP, TIF, MOV, WMV, WMA, MP3 files per month.
- Less than five JPG files over 500Kb and five Documents over 1Mb per month.
- Less than two uncompressed low-value files per month.
- Less than 10% overall growth of the Public data volume per month.
- Less than five new files per author per month.
Method of measure:
A global search of the Public Drive will be conducted for all BMP, TIF, MOV, WMV, WMA, MP3 and JPG files. All BMP and TIFF files will be flagged for attention. All MOV, WMV, WMA and MP3 files are flagged for removal. JPG files are sorted by size and any found to be over 500Kb are referred to the creator for reformatting.
A global search for all other file types will be conducted, sorted by date and author. They are then checked for relevance, size and volume by individual authors/creators.
Logical data Structure
Measure of performance:
- No files to be found in the wrong location.
Measure of conformance:
- Less than five files found to be in the wrong location per month.
Method of measure:
An on-going review of the contents of the Public drive will ensure that all files are placed in the most logical location. The relevance to a specific location will be assessed on the grounds of the content’s intended audience and the author’s role. Consultation with the author and author’s CO will determine the location of indistinct content.
Responsibility
Measure of performance:
- All files are named in accordance with Policy Statement.
- All files have author/creator information according to Policy Statement.
- All files to be located correctly according to the Region/Site/Unit structure.
Measure of conformance:
- Less than five files found to be incorrectly named per month.
- Less than five files found to be without correct author/creator identification per month.
- Less than five files found to be in the wrong location per month.
Method of measure:
A global search of all files in date order will be conducted. All recent files will be assessed as to being suitably named and labelled (per Policy Statement).
Conformance Enforcement:
The author/creator of large files (images over 500Kb and Documents over 1Mb) and/or a high number of new files (more than five new files per author) will be notified and required to reformat those files. This can be done by reducing the dimensions of screen-viewable images or converting DOC or EXL files to PDF. High numbers of new files must be justified by the author/creator.
The author/creator of inappropriate (inefficient) file formats (such as; BMP, TIF, MOV, WMV, WMA, MP3) will be notified and asked to remove or reformat them into JPG (in the case of photographic images). A business case is required for inclusion of any non-image media files (video and audio).
Any files that do not clearly identify their author/creator will be removed.
Authors/creators will be notified and asked to delete any duplicate file/s. It will be left to the discretion of the author as to which location the single instance will remain.
Users will be notified once of the need for their documents/files to be updated and resaved once they become older than six months; otherwise the files will be removed one week after notification.
DOC and EXL files may remain in their native form if they must be updated regularly. It will be recommended that all other reference documents are converted to the more efficient PDF format.
Sub-folders will be created for specific subjects, as required. Authors will be notified when a file is deemed to be in the wrong location.
An educational programme for efficient data storage practices will be included in all mandatory briefings. This will include archival techniques, a discussion on file formats as well as methods to identify low-value data.
References
John Reh, F. (2006). Benchmarking, Management About.com
http://management.about.com/cs/benchmarking/a/Benchmarking.htm
Arveson, P. (1998). What is a Balanced Scorecard? Balanced Scorecard.org
http://www.balancedscorecard.org/basics/bsc1.html
Results-Based Measurement: Better Measures in Less Time.
http://www.zigonperf.com/rbm.html