Very interesting! Have you had any success with automating this process any further?

Also, I would be very interested in any other findings regarding open source e-mail/data extraction applications. Have you found any other usable tools?

Thanks and with the best regards,
Martin

http://googleenterprise.blogspot.ro/2012/12/google-apps-vault-now-available-to.html?utm_source=entblog&utm_medium=blog&utm_campaign=Feed:+OfficialGoogleEnterpriseBlog+%28Official+Google+Enterprise+Blog%29

To see if you are eligible to purchase Google Vault, you can check the management dashboard on the right hand site which indicates which additional services can be purchased.

more from NPR (National Public Radio)

Laura
Unfollow
Laura Ewing-Moore • This is very helpful. Are you also planning to evaluate Nuix and/or Clearwell?
1 hour ago• Like

Jeffrey Goodwin • Our current evaluation is on a 450 Gig international Unicode data set. Our testing has consisted of iPro, eCatpure, and dtSearch. We have found that the ingestion rate of DT Search, is close to 150 Gig per day for a single workstation with an i7 Processor and 8 embedded CPU’s, 32 Gig of DRAM, and solid state Muskin drives.

We see that Nuix also has some comparable indexing features in their eDiscovery Platform. http://www.nuix.com/Nuix-eDiscovery-Platform, but we have not tested it yet, nor Clearwell’s eDiscovery suite (

For amount of data we were testing (450 Gig) we were able to Index all of the data in a few days, and the analytic results from the search keyword proposals ran very quickly thereafter. We were able to concatenate 133 term search term proposals in English, Simplified Chinese, and Traditional Chinese and make them fit within the 5000 character search term upper limit in dtSearch to obtain the total number of responsive documents to get a rough estimate on the number of documents. In dtSearch this breaks down to the individual message level within .PST’s, but responsive documents inside of an archive (e.g., .zip or .rar file) were reported as a single responsive document, and there is no deduplication in dtSearch.

In parallel to this, by using custom developed scripts with Windows 7 native robocopy, we were able to separate the processing of all email into specific repository and determine the number of email duplicates, and impact on the number of responsive documents for email by using different cutoff dates using LAW pre-Discovery, and in a similar effort iPro eCapture.

Our methods seemed to stand up to the the scrutiny of the litigation process involving attorney’s, the Magistrate, and other ESI expert witnesses and therefore dtSearch seemed to do the job in a reasonable time frame and cost.

I’d be interested to hear more about your experience with Nuix and Clearwell. Note that the only unstructured data we were dealing with was a couple of difference ERP systems, which have yet to be searched.

Best regards,
Jeff