Home

Thursday, April 29, 2010

User Experience (UX) in Web Portals

Planning and executing information management projects accounts for a major portion of the overall effort. However, there is still quite a bit of time that is required to be spent on refining and communicating concepts to colleagues and to the business. Perceptions about why you are developing a portal or system and what it will look like have a tremendous amount of impact to the success of a project.

User Experience tends to be one of the facets of information system design that draws quite a bit of controversy because it is very subjective. Combining information architecture concepts, system capabilities, features, security models, business requirements, functional requirements, standards, and policies to develop a design will get you most of the way there. However, there are still some very volatile aspects to the design process; these include emotion and perception.

The measure of user experience of a system is highly subjective. People's attitudes, situations, level of knowledge, and other factors produce an emotional response every time the person uses the system. Although I have lots of past experiences, projects, and communications to draw from for guidance on how best address matters related to UX debates; I've decided that I really need to collect some factual information to keep in the toolbox. I think that behavior studies, polls, statistics, and other sources of research are required to build a solid fact base to support decision making. It is simply not good enough to speculate how people think and how people will respond to a particular system design.

I plan to invest some time in this topic and then reflect on my findings in subsequent posts.

Tuesday, April 27, 2010

Sample Sizing Estimate for SharePoint Server 2010 Search Environment

Below is a sample search storage estimate for environment with 500GB content databases. This was calculated by adding up the sum of MDF files. There are so many considerations for storage, and this is just a rough sketch of the storage aspects related to the Search Service Application. This does not consider base installation overhead nor does it factor in other SharePoint Service Applications (SSA), Office integrations, User Profiles, Usage Database, PowerPivot, etc..

1.
ContentDBSum = Measure the size on disk of MDF files. For this example, 500 GB.

2.
IndexTotalIndexSize = ContentDBSum * .035 or
17.5 GB
QueryComponentIndexSize = TotalIndexSize / Number of Index Partitions
StorageQueryComponentIndexSize * 3 single query component with space for index merging
QueryComponentIndexSize * 4 single query component with space for index repartioning or 70 GB

3.
TotalPropertyDBSize = ContentDBSum *.015 or
8 GB
TotalPropertyDBLogSize = ContentDBSum *.0031 or 2 GB

4.
TempDBSize = TempDBSizeContentDBSum * .00034 or 1 GB

5.
TotalCrawlDBSize = ContentDBSum *.046 or
23 GB
TotalCrawlDBLogSize = ContentDBSum *.011 or
6 GB
ContentDBSum *.011 = TempDBSize = 6 GB

6.
SearchAdminDBSize = # items in index * .3 or override as
5 GB
In this case, I don't know how many items there will be in the index so I am going to allocate 5GB.

7.
Backup Size = basically, add all of the database sizes together.

Reference

Microsoft Technet (2010). Performance and capacity test results and recommendations. Retrieved April 27, 2010 from http://technet.microsoft.com/en-us/library/ff608068(office.14).aspx.

Friday, April 23, 2010

Windows Server 2008 R2 Data Center VS Enterprise

Here is a quick overview when comparing Windows Server 2008 R2 Datacenter vs. Enterprise. Based on the licensing model for Windows Server 2008 R2, it may be more cost effective for many virtual infrastructure environments to license Datacenter Edition rather than Enterprise Edition. Furthermore, the hot pluggable hardware features of Datacenter Edition may be a key discriminator.


Role: No difference

Core Installation Features: No difference

Features: No difference

Technical Specifications: See Below

Datacenter
· hot add processors
· hot replace memory
· hot replace processors
· Virtual image use rights = Unlimted
· X64 sockets = 64

Enterprise
· No - hot add processors
· No - hot replace memory
· No - hot replace processors
· Virtual image use rights = Host plus 4VM
· X64 sockets = 8



Reference

Windows Server 2008 R2: Overview of Editions
http://www.microsoft.com/windowsserver2008/en/us/r2-editions-overview.aspx

Friday, April 09, 2010

Inconsistency in Information Management System Governance, Projects is Kryponite

I continue to revisit the thought that a SharePoint governance model (policies, procedures, standards, resources, etc.) can be undermined if an organization does not employ proper and consistent governance for all of its information management activities and systems.

Consistency is important because it level sets the way that people approach information management projects, including SharePoint projects. If some projects were to take a fly by the seat of your pants approach, disregarding governance concepts, it is bad because the resulting deliverables will be inconsistent from what the guiding principles that the governance model is supporting. What is worse is the sloppy result is the one that is achieved first, since it takes less time to do things when you circumvent processes.

Meanwhile, you might have some projects following the rules and progressing a bit slowly while the wild wild west ones are moving fast and crossing the finish line earlier. By the time the "good" project finishes, the wild wild west has already become the de facto standard. A precedent is set and now all of those "unnecessary" rules are perceived as pockets of resistance and momentum builds in avoiding the rules rather than following them. This is a vicious cycle and requires quite a lot of effort to reverse the damage. It is much easier if the executive sponsors can support the governance concepts, then promote and mandate accordingly.

The benefits of the formal approach to projects and adherence to information management governance is obvious to anybody who has experienced a complete project life cycle and then stuck around long enough to observe the after effects. This topic is very much philosophical; do you want to manage projects short-sighted to get the project completed or do you want to manage projects for the greater good, long term to support the guiding principles of the governance model?

I will draw an analogy. You live in the mountains and have a wood stove. You can spend cheap money buying an axe, then hurry to chop wood with it to meet an immediate need of filling the stove once. Or, you can invest more money on a chainsaw and then be able to provide wood for the stove, more efficiently, every time you need it moving forward. It is called return on investment. Sustainable efficiency and the ability to produce higher quality deliverables (which are easier to maintain) is the result of following a slow and steady approach to information management projects; taking the time to remain aligned with the principles of the governance plan.

Blog Archive

Followers