toby's Journal: say whut?

Cloud computing is emerging as the computational and storage paradigm for ultra-scale data and analytics. A cloud is a pool of virtualized, commodity computer resources accessible via a Web-based interface in conjunction with a highly automated mechanism for managing those resources. Upon this infrastructure, a computational framework brings analytics to the data, effectively implementing a key tenant of cloud computing - ingest data once, move data rarely, and re-use data often. Thus cloud enables the rapid configuration and allocation of vast computational resources while supporting massively parallel processing over petabytes of distributed data to deliver both agility and power at scale. Google, the progenitor of cloud, and others have proven the extraordinary capability of cloud for specific applications within certain well defined domains (e.g. web search/index). Within the Intelligence Community (IC), substantial cloud computing initiatives already underway are poised to do similarly, tailoring cloud for specific objectives. Considered more broadly however, as an enterprise wide platform for the IC, several aspects of cloud are in need of further research and development. Principal among these is the cloud analytic tool set. Analysts working high op-tempo missions down range tell us the most rapid time scale is not of the data or the mission, rather it is the frequency with which they must change the way they view and manipulate the data. For these warfighters, sophisticated tools built nationally are obsolete the day they get deployed. What they clamor for are basic, elemental tools that they can apply in diverse contexts and compose into workflows that help them do their job easier and faster. This SBIR seeks to develop an analytic tool set designed to operate within cloud computing infrastructure, on cloud data. In addition to being cloud-native, the set of tools should reflect a basis-set of functions that together span the range of analytic activity. Each tool should capture / enable a basic element of analytic activity while presenting minimal dependencies on other elements or on application specifics. By developing a complete set of cloud-native tools that may be used independently or composed into workflows to build analytic product, this SBIR aims to provide the warfighter with the means to harness the full power of cloud in the face of rapid change.

Let's play buzzword bingo...

21 the
16 of
14 cloud
10 and
9 to
9 data
8 a
6 for
5 is
5 set
5 they
5 analytic
4 tools
4 computing
4 that
4 are
4 in
3 on
3 computational
3 resources
3 or
3 with
3 rapid
3 tool
3 scale
2 this
2 SBIR
2 range
2 workflows
2 these
2 native
2 analytics
2 basic
2 an
2 do
2 infrastructure
2 activity
2 into
2 specific
2 change
2 while
2 IC
2 within
2 should
2 as
2 power
1 full
1 elemental
1 sophisticated
1 way
1 paradigm
1 enables
1 both
1 Upon
1 Intelligence
1 framework
1 independently
1 based
1 over
1 nationally
1 index
1 emerging
1 massively
1 functions
1 pool
1 Principal
1 developing
1 effectively
1 key
1 complete
1 parallel
1 compose
1 once
1 re
1 aims
1 further
1 enterprise
1 at
1 rather
1 application
1 tell
1 managing
1 development
1 deliver
1 agility
1 applications
1 may
1 computer
1 highly
1 move
1 addition
1 working
1 allocation
1 need
1 not
1 ultra
1 mission
1 Within
1 warfighter
1 interface

Hey, that's what I do

[ynotds]

(including word frequency counts)

But mostly on data generated by discrete systems with interesting emergent properties.

It also sounds a lot like Wolfram|Alpha.

And I have an extremely relevant domain registration which I just need to live long enough to make use of.

As science was to the last half millennium, so analysis will need be to the next if we are to navigate emergent black swans and unknown unknowns.