Google Search Appliances (Archived Report)










PDF version of this report
You must have Adobe Acrobat reader to view, save, or print PDF files. The reader
is available for free
download
.

Archived Report
Google Search
Appliances

by Michael
Gariffo

Docid: 00011283

Publication Date: 1908

Report Type: PRODUCT

Preview

Google offered its Search Appliances to enable the retrieval of
information from enterprise storage and networks. These tools, based on
Google’s market-leading search engine, were designed to provide high
search quality, access control, and connectivity. As of 2018, Google
ceased selling new licenses for its Google Search Appliance product, but
it announced plans to continue to support existing customers through the
remainder of their license agreements, into 2019.

Report Contents:

Description

[return to top of this report]

Related
Faulkner Reports

Google Company Profile Vendor

Google Search Appliances were tools that reached deep
into enterprises’ storage and networks to retrieve any type of document or
data. With an interface similar to Google’s Internet search
engine, Google Search Appliance allows end users to submit
queries that can probe millions of internal and external sources – including file shares,
databases, intranets, Web servers, portals, real-time data in business
applications, and content management systems.


Vendor
Name: Google Inc.
Headquarters
1600 Amphitheatre Parkway
Mountain View CA 94043
Phone: (650) 623-4000
Fax: (650) 618-1499
E-Mail: info@google.com
Web: http://www.google.com
Type of Vendor: Internet Search Engine
Founded: 1998
Service Areas: Global
Stock Symbol: GOOG

Google Commerce Search

Google Commerce Search was targeted at enhancing online retail experiences. As
a hosted offering, Commerce Search was designed to handle increased seasonal
query volume. A merchandising dashboard allowed retailers to customize search result
rankings while also filtering options and promotions. With simple facet rules,
they could control how the left-hand navigation panel appears and could
automatically generate groupings based on the user’s query and matching items or
specify which attribute groups were shown to the user and in what order.
Administrators could customize search results and integrate them into other
applications by using the XML API. The XML API allowed users to access raw search
results for a query in XML format, including the URL and description text. This
allows retailers to view custom reports including result sets, top queries,
product promotion clicks, and special feature usage. Administrators can use
integrated Google Analytics to study how visitors are using an online store.

Figure 1. Commerce Search Merchandising Dashboard

Figure 1. Commerce Search Merchandising Dashboard

Source: Google

Google Search Appliance 7.6

The final version of Google Search Appliance was designed for large enterprises and
could
support billions of documents. It featured
direct database indexing; a feeds API allowing customers to push new content
into the index; supported more than 220 file types, including HTML,
Microsoft Office, PDF, PostScript, WordPerfect, and Lotus; RAID (to provide continued service in the event of a hard disk
failure); supported end-user-level document security, including a security
API; supported Cloud Connect, which links the appliance with Google Apps, Site Search,
and Twitter; and included People Search for examining individuals within an organization,
as well as dynamic navigation through metadata filters, and interoperability with
Microsoft Sharepoint. 

Features which were new to version 7 of the Google Search Appliance, which launched
in October 2012, included new relevance algorithm enhancements, document
previews, search result translations, entity recognition, Dynamic Navigation
improvements, the ability to moderate user-added results, connecter suite
enhancements, and improved language support.

The final update to the platform, version 7.6, which launched in early
2016, added support for document previews; dynamic navigation via filters such
as author, date, or location; the ability to group content into customizable collections,
and more. It also enhanced its existing support for entity
recognition, spellchecking, and language translation.

The physical component of the Google Search Appliance came in two models: the
G-100 and the G-500.

The G-100 was a rack-mounted single-unit
appliance that could be licensed to search up to 20 million documents. Additionally, multiple
G-100s could be linked together to support larger document counts, or to integrate searches across multiple departments, geographies, or
Web sites. Based on a RAID architecture and equipped with redundant components, the G100 provides built-in redundancy.
Figure 2 depicts the Google Search Appliance.

Figure 2. Google Search Appliance Model G100

Figure 2. Google Search Appliance Model G100

Source: Google

The G-500 could search up to 100 million documents. For larger deployments, multiple G-500 appliances can be linked together to search hundreds of millions, or even billions, of documents.
The G-500 enables built-in redundancy and failover, delivering maximum capacity and reliability.

End User Features

The Google Search Appliance offered end users many of the same benefits
they enjoyed with Google.com, including:

  1. Dynamic Navigation – A method of filtering search results by
    specific, user-defined metadata attributes. 
  2. Sidebar Elements – Allowing for the presentation of personal results
    from Apps Search, employee information from People Search, competitive
    intelligence from external Site Search, and real-time information from
    Twitter within a sidebar next to indexed content. 
  3. Self-Learning Scorer – A feature which used statistical regression to
    analyze and score specific links and user behaviors with the goal of making
    future searches more accurate. 
  4. Query Suggestions – Similar to Google Suggest, the search box
    would suggest query refinements
    to speed navigation.
  5. Social Search Features – With User-Added Results, end users across the enterprise
    could promote certain web pages for any given query.
  6. Personalized Search Experience – Allowed administrators to adjust search results for different user groups, based on department or function.
  7. Alerts – End users could subscribe to
    e-mail alerts for topics and documents of interest, choosing an hourly, daily, or weekly schedule.
  8. Automatic
    Spell check
  9. View as HTML – Displaying documents without needing the original client
    application (due to automatic reformatting of over 220 file formats into
    HTML).
  10. Sort by Date – Access to time-sensitive information first via date sorting.
  11. Date Range Search – Restricting search results to a particular date range.
  12. Advanced Boolean Search – Performing complex queries with
    more than 10 special query terms, including Boolean (AND, OR) searches.
  13. Related Queries – Define related queries for
    enterprise-specific acronyms or terminology, and have those terms displayed as suggested alternative
    queries.

Figure 3. Google Search Appliance Sample Search

Figure 3. Google Search Appliance Sample Search

Source: Google

Access Control Features

The Google Search Appliance provided document and
end-user-level access control across all enterprise content to ensure
end users only see search results for documents they have access to view. Access control elements include:

  1. Early Binding for Large Scale Deployments – Supported policy ACLs
    in the index, and enables early binding for even deployments ranging in the
    billions of documents.
  2. Universal Login – Enabled dynamically generated login pages to
    securely pass login info to back-end systems. 
  3. Multiple Cookie Domains – A feature which enabled single sign-on
    system to access content from multiple domains.
  4. SAML Batch Processing – Customers
    could customize security authentication and authorization using SAML, with SAML requests
    batch processed for more efficient authorization.
  5. Enterprise Single Sign-On
    – The Search Appliance integrated with LDAP, NTLM, Windows Integrated Authentication,
    and forms-based single sign-on security systems, including Oracle Access Manager and CA
    SiteMinder, to enable seamless secure searching.
  6. Kerberos Support – The Search Appliance provided
    native support for Kerberos, enabling a silent authentication experience for end users.
  7. X.509 Client Certificates – The Search Appliance served secure results to
    end users using X.509 client certificates.
  8. Secure Web Content – The Search Appliance enabled secure searching of information protected by basic HTTP authentication or NTLM.
  9. Content API
    End users could search across secure content using Google’s SAML Authorization API to integrate into existing access control systems.
  10. Lotus Domino
    Security –
    The Search Appliance integrated with Lotus Notes environments using fast, efficient crawling of Lotus Domino
    servers.

Applications

[return to top of this report]

Just as Google.com and other Internet search engines enable end users to
discover content of interest on the World Wide Web, the Google search appliances enabled enterprise end users to discover
content of interest within the realm of their own enterprise repositories.

In particular, the Google Search Appliance permitted large enterprises to
search millions, even billions, of documents distributed over disparate
databases. In addition to everyday searches, the Google Search Appliance
enables enterprises to:

  1. Locate intellectual property assets (thus helping to preserve
    enterprise investment).

  2. Scour databases for illicit content (such as copyrighted music or
    pornographic images).

  3. Gather data relevant to future sales and marketing initiatives.

  4. Analyze storage utilization (in an effort to inhibit out-of-control
    storage growth).

  5. Recover records in response to a subpoena or other legal or regulatory discovery
    order.

Environment

[return to top of this report]

To affect a virtually universal search capability, the Google Search
Appliance supported an extensive collection of languages and file types, and
provides "content connectors" to a host of content repositories.

Table 1. Google Search Appliance Search Environment

Languages

File Types

Content Repositories

Arabic, Chinese (Traditional and Simplified), Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hebrew, Hungarian, Icelandic, Italian, Japanese, Korean, Latvian, Lithuanian, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, and
Turkish

220 file types including HTML, PDF, MS Office, and many more

EMC Documentum, EMC VNX, IBM FileNet,
LDAP, Microsoft SharePoint, OpenText Livelink, BEA AquaLogic, EMC Documentum eRoom,
Hummingbird, IBM Websphere, Lotus Notes/Domino, Oracle Content Server (Stellent),
and SAP KM

Support

[return to top of this report]

Software Support

The Google customer support site will be available 24/7 to the remaining
grandfathered customers until the end of their current licensing agreement in
2019. While it operation, it provided a

range of help options including discussion forums, support
documentation, the latest software updates, and release notes. For
Site Search, support was included with the purchase price, and is delivered
through an online model. The support team was available via email and comprises engineers who
specialize in Google’s enterprise search
technologies, according to the vendor.

Hardware Replacement

If a Google Search Appliance failed during the term of a
customer’s support contract, Google furnished the customer with a new appliance.

Pricing

[return to top of this report]

New Google Search Appliance is no longer sold.

[return to top of this report]

About the Author

[return to top of this report]

Michael Gariffo is an editor for Faulkner Information Services. He
tracks and writes about enterprise software and the IT services sector, as well
as telecommunications and data networking.

[return to top of this report]