Text: THE PROBLEM : SEARCHING AND SHARING DATA
Searching for, and securely sharing , your data, in multiple formats ,
across platforms and between organizations and business
partners, is a difficult proposition. Data sharing has become a highpriority in many market sectors, but there are numerous practical
challenges to delivering on this difficult challen ge: disparate
database systems with differing schemas, structured (tabular
database information) versus unstructured data (Microsoft Word
documents, PDF files , web pages, etc.), varying security
requirements, and performance and cost considerations.
I HIGH DATA VOLUMES
DIG promotes a higher degree of data consumption when
compared with standard web sites or Portals. This is due to
the automation , and multi-faceted nature, of the federated
search . Giving users access to more of your data as well as
the data of other agencies, can boost the value of your data to
I INSTANT SOA
Visual Analytics' Digital Info rmation Gateway (DIG) solves this
problem. DIG provides the means by which you can search and
share your data, regardless of structure, location or format, within
your organization, or outside your organization , securely, efficiently
. • ,, • • L l
DIG : SEARCHING AND SHARING DATA SOLVED
How does DIG solve this challenge?
I HOMOGENIZED DATA
DIG provides a layer of "abstraction" for each data source
under its control , "mapping" a standard schema to the columns
of each shared source. When a user receives query results, all
columns returned adhere to this schema, regardless of the
name given to any column in the source database. Differences
between sources are hidden from the end-user, giving the enduser the illusion of querying a single, vast, database.
Wh at's Instant SOA? It's the instant transformation of your data
into a Services Oriented Architecture that can be leveraged by
anyone with access to the DIG network. DIG exposes shared
data sources as Web Services , allowing any web service client
to access the shared data. The DIG SOA framework provides a
robust set of Web Service AP ls that enable existing client
applications to easily use your data.
DIG is designed to access any type of data, and provides a
ready-made set of web services that are generic and that put
you in control. This is made possible by a modeling framework
that allows each data source to be mapped to the standard
schema. Once your data is controlled by a DIG server, that data
can be exposed as a web service - without any additional
VISUAL ANALYTICS INC.
I FEDERATED SEARCHING
I SINGLE POINT OF QUERY
DIG uses a distributed , clustered network approach to
executing searches. Multiple DIG servers distribute and
respond to queries, thus distributing workloads across servers ,
improving query response time for the end-user.
All of the data in a DIG network is automatically available
from a single query user interface. When a user enters a
query, the query is distributed across the DIG network,
executing against both structured and unstructured sources .
This same approach allows secure "outbound" sharing of data
between departments and to business partners. Any location
wishing to share data may install a DIG server that manages
the sharing activity. The data managed by each DIG server can
be shared , or restricted from sharing, as dictated by local
policy or as required by service agreements . Share a little, or
share a lot, or only share yes/no notifications of matches - all
of this is configured through a graphical user interface.
I SECURE SHARING
I SIMPLE TO QUERY
The user interface for the DIG query client is simple to use.
The interface resembles any number of Internet search
engines making it immediately understandable to the average
end user. DIG also supports a simple query language for
I WORD-LEVEL INDEXING
All the data managed by DIG can be word-level indexed to
speed performance . When a user queries for "John Smith ,"
this query is federated across participating DIG servers , and is
searched for in every column indexed by DIG. This allows a
user to find references to "John Smith" in dedicated columns
(like First_Name or Last_Name columns} , in a "Notes" or
"Comment" column , as well as in unstructured data.
Because the queries are federated , and because word-level
indexing happens for all columns in all shared sources, a
phrase like "John Smith" can even be found in a source where
no dedicated name fields exist.
All access to data is controlled at each DIG server and the
data streams between servers are encrypted. Any column of
a database can be shared, or not. DIG provides for both
user-level and group-level access at the data source , table
or column level. DIG also enables application-specific, rowlevel security to be applied through an open, plug-in
I VIRTUAL DATA WAREHOUSE AT YOUR
In essence, DIG provides a "virtual" data warehouse , without the
expense and work of creating a traditional , "real ," data
warehouse . DIG consolidates your data, without the need to
physically move the data from the source to a warehouse. This
approach also reduces the latency issues normally associated
with traditional data warehousing , in that a recurring ETL
(Extract , Transfer and Load) process must be applied to the
original source data to introduce it into the warehouse .
I A PICTURE IS WORTH
Below is a graphic displaying a common approach to
configuring a DIG network.
Document Path: ["1264-visual-analytics-brochure-dig.pdf"]