In its purest form, the Enterprise Service Bus (ESB) isn’t a technology but rather a model for inter-application communication. The reality is, however, that in the world of legacy enterprise application integration (EAI), heavyweight ESB technology forms the basis of a Service Oriented Architecture (SOA), representing a hub-and-spoke integration architecture to support a variety of point-to-point connections between service providers and consumers.
Used primarily for integrating services in heterogenous and complex application landscapes, the ESB was designed as a more agile and flexible model than client server. Over the years, however, it became common for companies and vendors to insert more and more business logic into the ESB, making it more complex, and laden with performance issues, especially evident when sending with large volumes of data over the bus. The shift to cloud applications in the enterprise (and the business requirements for fast cloud integration) and big data volumes, velocity and variety only further compounded the issues.
The problem with the traditional ESB is that its very fabric makes it incompatible with the nature of modern data. APIs have become the de facto method for interfacing to both enterprise and web systems – primarily using the REpresentational State Transfer (REST) protocol with the data encoded via JSON. The ESB primarily relies on Simple Object Assess Protocol (SOAP) and EXtensible Markup Language (XML) for data representation. From an integration platform perspective, REST and JSON are increasingly replacing SOAP and XML, making ESBs less relevant in the new SMACT (social, mobile, analytics, cloud and things) enterprise IT architecture.
Big data begs for agility, elasticity and high-speed throughput while dealing with large volumes of disparate multi-point data and transformations, both inside and outside the corporate firewall. Inherently tied to tightly coupled, complicated and ultimately bloated implementations which cost-prohibitive overhead, the ESB model and the legacy technology that support it begins to fracture under the pressure of modern data and ever evolving business requirements. Simply put, streaming data over a legacy ESB in the enterprise is no longer a viable option.
SnapLogic’s approach to the ESB rests on the continuous high-speed guaranteed delivery of data from disparate sources, both on-premises and in the cloud, while allowing in-stream filtering and transformations. Designed from the ground up for modern cloud application and big data requirements, SnapLogic’s technical foundation is built firmly on a 100% REST architecture and JSON lightweight data-interchange format.
SnapLogic Ultra Pipelines are always-active, providing a paradigm of continuous document ingestion with up to 10x acceleration in data delivery and turnaround rates, designed for the velocity requirements demands by today’s enterprise.
SnapLogic’s Elastic Integration Platform provides a rich set of functionality including an intuitive, cloud-based, drag-and-drop graphical integration orchestration console. Features of the platform include Ultra Pipelines (data flow orchestrations), pipeline versioning, aggregation, joins, unions, splitting, slowly changing dimensions (SCDs) and scheduling. Automatic monitoring of performance and requests ensures adherence to corporate governance policies and SLAs.
Removing the complexity and long, costly developmental cycles and management overhead so commonly associated with the legacy ESB technology, SnapLogic pipelines can be quickly configured by “citizen integrators” without advanced middleware skills or special training / experience. SnapLogic elastically scales up or down to accommodate workload demand using collections of JVMs (called Snaplexes) that can reside on-premise and/or in the cloud.
Because of dramatic evolutionary shifts in the volume, velocity and variety of data and the granularity of microservices, legacy ESB technology can no longer effectively deliver on its original promise. As the enterprise continues to migrate more toward microservices (complex applications broken down into small, independent processes) even more pressure is exerted downward on traditional ESB solutions, moving them further into the sunset.
SnapLogic’s concept of the ESB delivers the ultimate solution for flexibility, agility and elasticity. With quick configuration and setup, low-cost burdenless change management and the ability to drive and orchestrate data to and from multiple endpoints at great speed, EMA Research has noted that, “Even as pre-built Snaps eliminate work cycles and manual processes related to data integration, SnapLogic Ultra Pipelines accelerate data delivery. This parallel approach to addressing the twin challenges of big data integration and real-time processing reinforces SnapLogic’s position as a leader in the data integration and iPaaS space.”