{"id":4037,"date":"2017-09-21T15:22:15","date_gmt":"2017-09-21T22:22:15","guid":{"rendered":"http:\/\/www.couchbase.com\/blog\/?p=4037"},"modified":"2025-06-13T17:23:13","modified_gmt":"2025-06-14T00:23:13","slug":"gear-n1ql-era-engagement-databases","status":"publish","type":"post","link":"https:\/\/www.couchbase.com\/blog\/es\/gear-n1ql-era-engagement-databases\/","title":{"rendered":"Prep\u00e1rese con N1QL para la era de las bases de datos de compromiso"},"content":{"rendered":"

Couchbase Engagement Database Platform<\/span><\/strong><\/b><\/h2>\n

Woo-Hoo!!! Have you heard of Engagement Databases<\/a>? I am sure, \u201cNo\u201d, not until Couchbase announced<\/a> it with 5.0 release. In today\u2019s digital economy, success of a business is dependent on its ability to leverage successful digital transformation to provide superior customer engagement & experience. DNA of all this lies in data, and that calls for a powerful database platform which can tackle the data problems in Systems of Engagement at scale, at ease, and at the roots. These problems are pretty unique and hence the solutions should be too. Superior understanding of the problems in this space and elegant & efficient solutions to those problems led to the discovery of a new category of databases, we call <\/span>Engagement Databases. Such a database is complementary to the Transactional and Analytical databases you might have in your data fabric infrastructure. Following picture shows how they fit together.<\/span><\/strong><\/b><\/p>\n

<\/p>\n

N1QL<\/a><\/b>is the SQL extended query language designed to enable your applications leverage the power of Engagement Databases & JSON. N1QL has all sophistication and bells & whistles necessary to deal with flexible schemas and the hierarchically structured JSON data. You can write simple declarative queries using N1QL to naturally deal with such data.<\/p>\n

Couchbase 5.0 brings a number of rich query functionality and performance optimizations in N1QL to meet the demands of customer engagement focussed applications. Many of the functionality and performance improvements also include corresponding enhancements in GSI Indexes to help N1QL fully leverage the optimizations. See Couchbase Server 5.0 what\u2019s new<\/a> for complete list and details.<\/span><\/strong><\/b><\/p>\n

Adaptive Indexes<\/span><\/strong><\/b><\/h1>\n

Adaptive Indexes are special type of indexes in Couchbase that can index all or a set of specified fields of a document in extremely adaptable way. Unlike composite GSI indexes, such an index is generic in nature. Let me explain.. Adaptive Indexes can efficiently lookup any of the index-key values without restrictions such as prefix index-key ordering, in typical composite indexes. This enables efficient ad-hoc search queries that can use WHERE clause predicates on arbitrary fields without requiring to create multiple composite indexes or different combinations of index keys. <\/span><\/strong><\/b><\/p>\n

For example, consider a user profile use case. A person’s profile may need to be searched based on any of the personal attributes such as first name, last name, age, city, address, job, title, company, foaf, etc. Similarly, a hotel room availability may be searched based on wide criteria, such as room facilities, amenities, price, location, distance from specific points of interest etc. In such scenarios, traditional secondary composite indexes can’t be used effectively (see section Contrast with Composite Indexes<\/a> to understand the issues). Adaptive indexes solve this problem by creating one generic index that can run such ad hoc search queries effectively and efficiently.<\/span><\/strong><\/b><\/p>\n

For example, following figure(1) depicts how three different indexes I1, I2, I3 are needed for the 3 queries Q1, Q2, Q3. Further, a query on any other non-indexed fields is not possible without creating respective indexes. Moreover, \u00a0a query on just age, such as Q4, will not be able to use the index I2, as it needs the prefix index-key id aswell. Figure(2) shows how a single adaptive index can help queries on all fields irrespective of the index-key order.<\/span><\/strong><\/b><\/p>\n

<\/p>\n

<\/p>\n

Note that, adaptive indexes is not a panacea. These figures may give an impression that adaptive indexes are better than simple\/composite indexes. It is not entirely true. If you know queries (and index-keys) apriori, then it is always better to use normal secondary indexes. Because, they would be faster than adaptive indexes. Remember that, adaptive indexes can be much bigger in size as they index multiple\/all index keys and corresponding values. So, they may not perform as fast as simple secondary indexes. However, adaptive indexes solve the problems that secondary indexes can\u2019t with little overhead. <\/span><\/strong><\/b><\/p>\n

Life is all about trade-offs!! For more details and examples, refer to the Adaptive Index<\/a> documentation.<\/span><\/strong><\/b><\/p>\n

Powerful Subquery expressions<\/span><\/h1>\n

N1QL is a powerful query language. It is SQL for JSON, and it has all necessary language extensions and expressions to deal with many subtleties in the JSON data modeling, such as:<\/span><\/p>\n