[EHPweb] Earthquake Hazards Program Web System Architecture proposal document

[Jill McCarthy]

Lisa and Web Team,

Nicely done!  I have a few comments to share.

Most notably, I still don't quite get how you can avoid the need to 
replicate between the two Master servers?   Sounds like you're operating 
the two nodes separately and assuming they will build identical pages. Are 
there realistic examples where this strategy might fail, and for whatever 
reason, you end up with contrasting content on the two nodes? 

Jill

                                                    Jill McCarthy
                                                    USGS
                                                    Geologic Hazards 
Science Center
                                                    Golden, CO, USA

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
jmccarthy at usgs.gov                           P.O. Box 25046, MS-966
Office:  303-273-8579                             Denver Federal Center
FAX:     303-273-8583                             Denver, CO 80225, USA
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



From:
Lisa A Wald <lisa at usgs.gov>
To:
EHPweb <EHPweb at geohazards.usgs.gov>
Cc:
Linda Pratt <lkpratt at usgs.gov>, Jill McCarthy <jmccarthy at usgs.gov>, Tom 
Brocher <brocher at usgs.gov>, Bill Leith <wleith at usgs.gov>, Michael Blanpied 
<mblanpied at usgs.gov>, David Applegate <applegate at usgs.gov>
Date:
10/15/2010 03:11 PM
Subject:
Earthquake Hazards Program Web System Architecture proposal document




The EHP Web Team has been researching and discussing the next evolution of 
our Web System Architecture for over a year now, and we have finally 
created a document with our proposed configuration.  This system 
architecture is forward-looking and extensible to include more hardware if 
needed in the future, and other systems such as the Earthquake 
Notification Service, ShakeMap, PAGER, etc.

The growing complexity and volume of our website, as well as the 
popularity (approx. 28 million page-views per week), and the requirement 
for guaranteed availability calls for a more robust and reliable system 
than we have now, and we believe that this proposed solution will serve 
the needs for the foreseeable future.

The authors and consultants for this architecture include Eric Martinez, 
Jeremy Fee, Ray Buland, Chris Bidwell, Stan Schwarz, Judy Tegeler, and 
many other product providers.

Please provide us with feedback and comments by October 31 by replying to 
this email.  You may reply to the EHPWeb email list or just to me.

Thanks!
- Lisa
------

[attachment "EHP System Architecture.pdf" deleted by Jill 
McCarthy/GD/USGS/DOI] 



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geohazards.usgs.gov/pipermail/ehpweb/attachments/20101016/7a09c4fe/attachment-0001.html>