Post reply

Databases for Scientific/Environmental Data Management

  • December 18, 2007 at 7:40 am

    #93873

    Hello -

    As I read these posts, I find a lot of what I am interested in, but the focus is often different from what I am developing toward in my career.

    Is there anybody out there that, like me, is looking at database administration/developement in the context of maintaining large, multi-year or multi-site datasets for the environmental sciences? I think our questions and focus are somewhat different. My questions are the following:

    1) What resources have you found helpful?

    2) Do you have any certifications/degrees that have been helpful in you development as a DBA for scientific data?

    3) Is there a forum like this - OR, could we create a sub-forum here?? - that is dedicated to the specific questions for scientific database management?

    Thanks,

    SA

  • December 18, 2007 at 8:05 am

    #761039

    I am not sure what you mean by maintaining a database for environmental science, but I am of the opinion that regardless of the data being stored in the database the fundamental skills are the same which would be:

    1. Understand the basics of database design - normalization, indexing

    2. Understand the data being stored

    You can both of these through this site and other similar sites and books like the Inside Sql Server series. I am sure there are others with more and better recommendations than myself, but this is a start.

    Jack Corbett
    Consultant - Straight Path Solutions
    Check out these links on how to get faster and more accurate answers:
    Forum Etiquette: How to post data/code on a forum to get the best help
    Need an Answer? Actually, No ... You Need a Question

  • December 18, 2007 at 8:10 am

    #761041

    I tend to agree with Jack that any database could work. However I'll qualify in terms of what scale/rate you're looking at. If you were to want to store high volumes of real time data, like multiple water pressures occuring every second, then you might look at something like StreamBase (www.streambase.com), as a database engine suited to processing streams of data.

    If you are looking for large volumes of storage, segregation by year, etc., something that would do well with partitions, then I think Oracle, SQL Server, DB2, would all work. MySQL, PostgreSQL as well.

    If you need lots of people collaborating across a large number of nodes, you might need Oracle Parallel server, but that would be extreme.

  • December 18, 2007 at 8:24 am

    #761054

    Thanks for the replies-

    I definitely agree that the basics of database management - integrity through normalization, understanding the data - are going to get me through the bulk of what I need to do on the data side. And, yes, all "industrial sized" RDBMS would work - perhaps even Access would work (even large databases for us don't come close to the size in many business applications).

    But I guess I am looking more for the resources for, and prevalence of, "scientists" (sometime I use that word lightly to describe myself) administering their databases.

    I can't describe the specific situations, but I feel there is something slightly different - potentially just epistemologically, and perhaps with the actual approaches we would use - when dealing with environmental data, interacting with our users - scientists, stakeholders, clients - when there are research-based questions involved.

    So, I guess my post was a search for anybody out there doing that.

    Thanks,

    SA

  • December 18, 2007 at 9:15 am

    #761091

    Hi,

    I am not a scientist but I have worked on a number of database and reporting applications for an environmental services company in the UK. I have found that there are not that many differences in the way in which these applications work when compared to other industries I have worked in(i.e transport, food services, aerospace, and military).

    I understand that you can't give specifics at the moment, but I am sure that any post in the forums relating to a RDBMS will get adequate responses; even if the terminology used is different. If a plain-english explanation can be given by the originator then the respondees should be able to translate.


    Ade

    A Freudian Slip is when you say one thing and mean your mother.
    For detail-enriched answers, ask detail-enriched questions...[/url]

  • December 18, 2007 at 11:53 am

    #761181

    A DBA and a scientist is two different jobs. I worked in many different companies doing totally different things. They hired me for my SQL Server skill even I did not know their business.

    If you want to build a database for scientific research, just tell the DBA what you want to store. Am I missing something here? Am I too naive?

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply