Glossary of Terms

Term Definition
Alteration A change in the Destination data type to accommodate variations in the Source data type to avoid sidelining of Events for data type mismatches.
Auto Mapping A feature in Hevo that automatically maps new Event Types and their fields to the existing Event Types and their fields in the Destination.
BinLog A set of log files in MySQL that record every statement that is executed (adding, deleting, updating records or tables). The log files have a retention period after which they are deleted by the MySQL server. BinLog, short for Binary Log, can be selected as a Pipeline mode for MySQL Sources.
Bookkeeping The process applied by Hevo Activate on the data fetched from the Warehouse to derive the differences over two successive Activation runs before synchronizing these with the Target object.
Candidate Key Any key or group of keys that uniquely identifies the rows in a table. The candidate key defines which keys can form the primary key for that table. A table can have multiple candidate keys.
For example:
- Single key: The teacher’s employee ID is a candidate key for the faculty table in the students’ database.
- Group of keys: The teacher’s name and the subject taught by the teacher together form a candidate key for the subjects table in the students’ database.
Collection A table in MongoDB.
Connector A general term for a system that Hevo integrates with another service/software. In Hevo, connectors are Sources, for example, MySQL or Google Ads.
Data Type The form that data exists in. Each field or value has a data type, for example, document, scalar, set, string or integer.
Deduplication The act of removing duplicates from the replicated data.
Destination A database, file system, or data warehouse that acts as the endpoint of a data Pipeline, and into which, the data from the Source is finally loaded.
Document A structure that contains other primitive or complex data types. This is usually in a JSON structure. For example, a record or a field representing all the profile properties of a user.
Document database A type of non-relational database designed to store and query data as JSON-like documents. A document is a record in a Document Database.
Drop (table) The act of deleting all the data from a table (in the Destination system). This is a permanent action. Hevo does not delete the table itself; users have an option to do so if required.
Event A fundamental unit of data that represents the creation, update, or deletion of information in the Source and can be replicated to a Destination system. For example, a new document in a MongoDB collection, or updated contact details in the Contacts table in MySQL.
Event Type Groups of Events created based on the entity they are ingested from at the Source. For example, a table Customers in your Shopify data can be considered an Event Type, and all Events ingested from this table fall under this type.
Granularity The level of detail in a data set. For example, Daily is the lowest time granularity that you can have on Bing Ads API.
Immutable data The data that cannot be modified after creation. For example, the values assigned to a string in Java.
Incremental Data The new or modified data in the Source post-Pipeline creation.
Indexes Special lookup tables that the database search engine can use to speed up data retrieval. An index is like a pointer to the data in a table.
Ingestion The act of retrieving or fetching data from the Source.
Instance The data stored in a database at a particular moment in time.
Keys A single column or a group of columns that can uniquely identify rows (or tuples) in a table.
Latency The time it takes to load the data into the Destination once it is ingested from the Source. Also called data latency or end-to-end latency.
Mutable data The data that can be modified after creation. Everything except a String type value is mutable by default in Java. For example, the elements in an array can be assigned new values after initialization.
Non-nullable columns Table columns that do not accept NULL values. This forces a field to always contain a value, which means that you cannot insert a new record or update an existing record, if this column field does not contain a value.
Offset A value in the latest record fetched. Offset is used to identify the starting point for the next set of results to return.
OpLog A collection present in MongoDB that keeps a record of all the operations that modify the data stored in the database.
Parallelism A process in which when multiple tasks run at the same time. For example, an application can split its tasks up into smaller subtasks that can be processed in parallel, for instance on multiple CPUs at the same time.
Payload The essential information sent or received with the HTTP methods such as GET or POST, in an API.
For example, in the following JSON response:

Payload Example

The payload is the message, ‘Welcome, world!”.
Pipeline A pre-defined framework of user-configured processes that move data from one system to another, typically with Transformations that make it easier to analyze the data.
Pipeline Frequency The frequency at which a Pipeline ingests data from its Source. Read Scheduling a Pipeline.
Poll-based ingestion The process by which the data is read from the Source periodically, depending on a set schedule.
Primary Key A primary key is a non-null candidate key selected to uniquely identify every row in that table. A primary key is a candidate key but the reverse may not be true.
For example, the roll number of a student in the students’ marks database uniquely identifies a student.
Primitive Data Type A data type pre-defined by the programming language. For example, int, long, float, and double in Java.
Private app An app built by a company for its internal use; not listed on the Intercom App Store.
For example, ChargeDesk built a private Messenger app for their customer support team to easily share invoices with their customers.
Public app An app listed on the Intercom App Store built by a company for use by other companies.
For example, Aircall, a cloud-based phone system, is a public app.
Push-based ingestion The process by which Hevo acts as a receiver and the Source holds the responsibility to send/post data to Hevo. This applies to webhook-based Sources.
Range Attributes Values used to specify the minimum and maximum limit within which a field’s value may lie.
Rate Limits The limits imposed by an API vendor, such as Intercom, HubSpot, on the number of API requests sent to their public APIs. Rate limits only apply on calls to the REST API, applications using OAuth, and Integrations using API keys.
SaaS (Software as a Service) A software distribution model where the application is hosted by a company on its servers and is accessed by clients via the internet by paying a subscription fee. For example, Salesforce.
Scalar A simple, primitive data type with values like a number or text. For example, int, bool and character in Java.
Schema The metadata of an Event, comprising details like its name, fields within the Event, and the data types of these fields.
Schema Mapping A mapping between fields in the Source schema and the fields of the Destination Schema. It determines the columns in the Destination table into which the incoming data of each Source field must be loaded.
Secure Shell (SSH) A network protocol that is used to create a secure channel over a network between a client and server application to transfer data.
Service Account A Google account associated with a team rather than an individual user. It is user-independent and requires a Key to authenticate the connection. Service accounts are used when workloads are run on virtual machines.
Set An interface that implements the mathematical set. A set contains no duplicate elements.
Sidelining When Events are not loaded in the Destination table due to mismatching of data type with the Source table for the same column.
Sink A synonym for Destination. See Destinations.
Source The application or database from which Hevo ingests the data. For example, GitHub, Facebook Pages, and MySQL.
Target Any CRM application to which data may be loaded using Hevo Activate. For example, HubSpot, Salesforce.
Target object The table in your CRM applications such as Salesforce and HubSpot with which data is synchronized by an Activation.
Transformation The process of changing the format and structure of data. Read Transformations.
Warehouse A place to store the data accumulated from a wide range of heterogeneous Sources, generally used for data analysis and reporting. For example, Amazon Redshift, Google BigQuery or Snowflake.
Workspace A space created in an Intercom account to store users’ data. Users can create different workspaces within an account to organize the data based on their functionality, such as Support, Customer Onboarding, Lead Generation, or Customer Engagement.
Write Ahead Logs (WAL) A logging mechanism used by PostgreSQL. It is a log that maintains transactions that take place in the database. So, even the smallest change in data is written to this log before it is applied to the PostgreSQL database.
Last updated on 09 Nov 2021