The essential feature of an R&D Data Cloud is that it gathers all your data in one place: enriching it with metadata, tags, labels, etc., and parsing and harmonizing it using consistent schemas. That’s what gives Tetra Data Platform power to act as a single source of truth: aligning self-similar data so you can compare and analyze it.
The trick is making that both powerful and simple. The 3.1 release of Tetra Data Platform debuts a reorganized search interface sporting (among other improvements) a Google-like full-text search feature that makes it fast, easy, and intuitive to query the data lake, explore and download results, and/or quickly iterate: useful, for example, in developing performant queries for TDP’s more-technical search tools.
By default, the search screen comes up ready to accept typed queries in TDP’s standard search syntax, based on ElasticSearch’s built-in query_string search capability (a good balance of simple and powerful, but sometimes daunting to neophytes). Full-text search can be turned on by flipping a toggle on the search screen (see Illustration 1), then typing into the search box. You can find full, partial, or approximate matches, leveraging your “Google-fu” directly to explore the TDP data lake. Searches can also be refined with Filters, applied using a popdown Options dialog (see Illustration 2).
Matched results are displayed below (by default), with matches highlighted. You can drill down into returned data items directly, examining schema and values, download search results in .csv format, etc. The new release even lets you preview common stored-data formats (e.g., .pdf) right in the webUI. So you have everything you need to get work done, fast.
As you work, your search phrases and filters are combined in an ElasticSearch Query Language query, which you can review, modify, and re-execute directly (see Illustration 3). All TDP search tools — from the simplest to the most powerful — are thus now linked together to enable a natural workflow: useful for ad-hoc checks, quickly browsing and extracting records, learning the more-technical methods of searching, or for developing well-formed queries for programmatic application via the API.
For more on TDP’s growing system of search features, please see our documentation.
Multiple enhancements to search and new EQL affordances make TDP easier and faster for scientists and developers to use. These improvements are complemented by several improvements to quality-of-life for TDP administrators/operators.
Cloud Configuration: We want TDP to deliver value quickly, which means connecting instrument connectors to agents efficiently. WIth release 3.1, admins can now select a connector and create its agent in one place, instead of going elsewhere (i.e., the Data Hub, Data Management or User-Defined Integration screens) to do so. In the same screen, admins can also view agent status and details. Fewer clicks and context-switches lets you wire things up faster, with fewer mis-steps.
System-wide health/status display: Administrators (and users) can now view the health and performance of TDP components and pipelines from a set of tabbed dashboards. Realtime updates means you can easily locate and intervene if agents, connectors, or data hubs are stressed or offline; and quickly ascertain the cause of pipelines running slowly.
TDP version 3.1 updates aren’t all profound — some are just skin-deep, like a newly-refreshed webUI design with contemporary branding and improved information layout. Our documentation has also been improved, both cosmetically and in terms of discoverability: finding information has never been simpler.
We encourage TDP users to update as soon as practical, to take advantage of new search and operations features (plus bug fixes). Among the latter, version 3.1 of TDP updates npm http-proxy-agent, recently discovered to have a vulnerability to man-in-the-middle attacks.