On average, most businesses currently double the number of digital documents they have every twelve to eighteen months. The impact of this rapid addition of content, argues Google, is that the search functionality of an organization’s databases and Websites need to be scalable in a dynamic environment. Google maintains that scalability is a crucial need of enterprises today, which is why the new version of Google’s Search Appliance (GSA) for enterprise search customers has added a powerful, dynamic scalability feature, allowing businesses to now index billions of documents, in addition to indexing Web pages.
To extend search into the enterprise, Google offers businesses its GSA product, a yellow box (that resembles a slice of Swiss cheese), which is based on a standard Dell server and is powered by Xeon 5500 Series processors from Intel. The GSA can index any enterprise data generated by Oracle databases, SAP systems, Documentum, SharePoint, Salesforce.com, HR systems, intranets, wikis, and more, and presents it to employees in a familiar Google-like interface. The sixth generation GSA will be able to index 30 million documents, compared to 10 million in the last generation. In the sixth generation of the GSA, Google is now adding functionality to let businesses stack GSAs on top of each other, so businesses can search billions of documents across integrated GSAs. The fifth generation GSAs were not able to be interconnected.
The new version of the GSA also allows businesses to separate data by sector of an organization (engineering, marketing, finance) but still be able to provide unified search results of all data contained within each appliance. Each department can also monitor and regulate which documents should be integrated into enterprise-wide search and which should be kept within the department’s search.
In the new version of the GSA, Google has upgraded the customization of the appliance, adding several new features to help businesses tailor the GSA to their search needs. First, Google now allows administrators to specify whether the GSA will implement late binding, which is real time authorization of whether a user has access to a document, or early binding, where the GSA holds a cache of existing policy information about who can access which documents. Second, if a business has multiple GSAs operating search, administrators can give more importance in search results to a particular appliance (engineering documents vs. marketing documents). Developers can also add an extra ranking framework to stack search results.
Google is also revving up search results in the GSA, adding social search features such as suggested search and user-added results, that aggregate knowledge across the organization for more precise search results. Additionally, users can now enable cross-language search, where the GSA will translate their search results in real-time into whichever language they choose.
Google now counts 25,000 enterprise search customers, up from last year’s 20,000 customers. Over half of customers use Google’s search appliance and the rest use its hosted site search and other enterprise products. Last year, Google added results based on personalization and Google Alerts functionality to the GSA. This year’s emphasis on scale and customization reinforces Google’s potentially strong enterprise strategy for both content inside the firewall and in the cloud.