Too much of a good thing?
Sep. 6 2016 — 8:59a.m.
DYNAMIC PAGE -- HIGHEST POSSIBLE CLASSIFICATION IS TOP SECRET // SI / TK // REL TO USA AUS CAN GBR NZL (U) Too Much of a Good Thing? FROM: Customer Outreach and Advocacy, Menwith Hill Station (F77) Unknown Run Date: 08/10/2006 (S//SI) Menwith Hill's COUNTRYSTORE project looks for ways to make the sorting, storage and retrieval of data more efficient and effective. (S//SI) Just over a decade ago, most folks within the intelligence community believed we would suffer from a dearth of signal data from which to extract SIGINT. They thought the use of fiber optic cabling would fully replace satellite communications, thus making the intercept of our high-threat targets more problematic. This, coupled with the increased use of sophisticated, commercially available cryptography, was viewed as the "beginning of the end" of easily exploited target communication paths. (S//SI) Flash forward to where we find ourselves today. The commercial satellite communication business is alive and well and bursting at the seams with increasingly sophisticated bulk DNI (Digital Network Intelligence) traffic that is largely unencrypted. This data source alone provides more data for Menwith Hill analysts to sift through than our entire enterprise had to deal with in the not-so-distant past. To twist a phrase, it would appear that communication begets communication. (S//SI) What's the problem, you ask? More data is a good thing, right? Well, anyone who has ever spent time on search engines like Google¿ might get a hint at the problem. Accessing too much data does not necessarily lead to good information upon which to make good decisions (or in our case, good SIGINT) unless you know what to keep and what to throw out. (S//SI) How do you sort through the gigabytes of data that we collect on a daily basis and figure out the best bits to keep that will lead you to better understand or learn about new targets of interest? We've proven we can't keep it all, and even if we did, we don't have all the tools and technology in place to help us intelligently sift through the data to support the various types of development that we would like to accomplish. So what's a forward-looking SIGDEV site (like Menwith Hill) to do? Using a few familiar colloquialisms, simply figure out a way to "separate the wheat from the chaff," while making sure you "don't throw the baby out with the bathwater." (S//SI) This is not a new problem for NSA. However, it is a problem that Menwith Hill has recently begun committing resources to in the form of its latest SKYLAB initiative known as COUNTRYSTORE. Although the problem is not new, the signals and data content within those signals are continuously changing as commercial communication vendors come up with new and innovative ways of pushing more data through satellites. (S//SI) The objective of the COUNTRYSTORE initiative is to work with a broad range of SIGDEV stakeholders and determine a set of "business rules" to thin out the data that we store at site. By doing this, it will allow us to apply subsequent analytics to this reduced set of data in a more efficient and effective manner. Once these business rules are discovered and quantified, the second phase of this initiative will be to codify them in software that will allow this intelligent data reduction to actually occur. (S//SI) For more information on this project, or if you are interested in helping, please contact either or . A core team of skilled individuals is being assembled to tackle this initiative. The team leads would like to hear any "great ideas" you might have on how to make this project a success. MHS
MHS (U//FOUO) Menwith Hill Station POC: (U//FOUO) This article is reprinted from Menwith Hill Station's Horizon newsletter, August edition.