SPEAR

Introducing SPEAR

SPEAR is the game-changing TeraHelix paradigm that accelerates the delivery of any data-driven initiative. This one-of-a-kind TeraHelix capability sits at the heart of the TeraHelix ecosystem and powers all our data platform accelerators. It is the technical realisation of our collective experience spanning decades in the forefront of financial services technology.

Spear logically structures enterprise data models to transcend implementation technologies.

With Spear, your enterprise’s greatest asset - its data and the intellectual property tied up in its structure - is no longer bound to a particular technology implementation choice.

Freedom to move between development platforms, disparate systems, cloud providers and storage implementations.

Enterprise data and intellectual property are no longer bound by your organisation's technology choices.

Flexible Data Representation

Spear seamlessly moves between different representations of the same data model.

Spear represents the same data model in the structure appropriate for each application – allowing collaboration between different user groups. For example, nested objects are preferred by quantitative and software developers, whereas relational or tabular formats are favored for data science and reporting purposes. The data format can vary from high-level text-based formats (e.g., XML, JSON) to optimised low-level on-the-wire formats (e.g., Google ProtoBuf) or analytical formats (e.g., Parquet, ORC)

Flexible Data Representation

Spear enables enterprises to unify disparate data models into a single 'canonical' representation.

The rate at which new technologies are being developed is only increasing. It is more important than ever for enterprises to standardise their data model and future-proof it against an constantly evolving technology landscape. The standardisation of data enables efficient reporting, aggregation, and machine learning processes. These standards facilitate increased automation and innovation.

A canonical model establishes one data representation and source of truth for the enterprise.

The crafting of multiple tactical conversions between different data models causes confusion between teams in the enterprise and holds back innovation.

Non-existent or inadequate data models drive users to duplicate and ‘augment’ data at various touch-points throughout the enterprise.

Errors and technical debt accumulate silently until critical, causing report inconsistencies, costly corrections, and data field confusion.

Spear provides a rich set of abstractions, features, and technical integrations, making it easier to work with the canonical model and 'do it right,' rather than succumbing to industrial-scale duplication and misleading shadow models.

Spear canonical models can be authored, where subject matter experts can define their models

They can also be generated through consuming existing data models, such as FpML or models from analytics libraries.

Starting from scratch can be daunting.

A better approach is sometimes to start with an existing model and build up from there. Spear definitions are rich enough to represent both industry standards and in-house models.

In authored models, you can leverage features such as randomisation for generation of test data and annotations for built-in validations.

Utilise the ability to generate data reporting models from regulator defined specifications.

Data professionals will find Spear intuitive and familiar.

Spear is laser-focused on serving as a data structuring language, disregarding general programming language constructs that are unnecessary from a data perspective.

Standard Data Definition Constructs

Spear includes standard data definition constructs. These standard constructs include primitives, user-defined enums, namespaces, interfaces, structures, inheritance, polymorphism, generic types, and keyed structures treated as first-class citizens.

Advanced Data Definition Constructs

Spear provides advanced data definition constructs like functors, functional interfaces, unions, UnionStructs, StructsDefs, PropStructs, PropInstances, and annotations for Aspect-Oriented Data Definitions. It also supports relational (rectangular) data definitions.

SPEAR Features - Runtime Stage

Spear's Comprehensive Runtime Data Tools and Features

Spear embodies best practices as standard, encompassing equality, data differencing/comparisons, completeness, immutability, introspection, reflection, type registries, coercion, and dynamic features. It also offers runtime type generation and compilation, 'soft' structures and relations, type conversions, dynamic object-to-relation capabilities, supporting libraries, automatic user interface generation, and automatic structure inference (e.g., Spear from CSV).

Being able to auto-generate integrations between your data model and underlying technology implementation is the only way to stay ahead of the curve.

Distinguish yourself with Spear's integration features.

Development Platforms include Java and JVM Languages (e.g., Scala, Clojure, and Kotlin), Python, TypeScript, and C# and .NET Languages.

Data Formats offer two-way interoperability, including textual formats like XML, JSON, YAML, and CSV, as well as binary formats like ProtoBuf and Linear Buffer.

Data and Storage Platform Support includes object stores like Apache HBase and MongoDB, relational stores with dynamic SQL from Spear (e.g., Apache Phoenix), and data processing engine support with Apache Spark.