Bharath Yadla, Vice President Product Strategy, Ecosystems for Aerospike Blog, Aerospike Products and Technology

Digital Transformation and Industry 4.0 have brought enormous amounts of pressure upon traditional enterprises looking to avoid being disrupted by their digitally native competitors. These traditional enterprises (also known as Digital Immigrants) have a path towards digital migration which is primarily driven by embracing strategies to build digital business models, which in turn require digital architectures and systems.

Aerospike Database – Core Foundation of Digital Architecture

Any digital architecture is as strong or as weak as the database that the architecture is built upon; thus, a database becomes a core and foundational block. Aerospike, for the past eight years, has been battle-tested by digitally native enterprises. These enterprises have built their consumer-scale digital architectures, which enjoy being highly performant while simultaneously scaling to accommodate data growth of 5, 10 and even 40x.

Aerospike has leapt forward into high-fidelity data uses with Aerospike 4.0 empowering customers with a database that has strong consistency combined with the uncompromising Aerospike hallmark of “unrivaled performance at scale.” Aerospike 4.0 turbocharges Digital Immigrants (traditional enterprises) with database features that support use cases which are not just systems of engagemen driven, but also use cases that are transaction-oriented where high-fidelity is a must.

Aerospike Ecosystems – Enabling Frictionless Enterprise Deployments

As we learned from our numerous deployments, it became clear that Digital Immigrants feel both the strength and burden of their existing systems and applications during their transformation journey, while Digital Natives enjoy the lightness of existing systems, but draw the force of implementing from the ground up. In both the cases, the need for frictionless deployment of a mission-critical database like Aerospike became clear. However, for Digital Immigrants (Enterprises), three distinctive areas are crucial

1. Interoperability: Ability to interoperate with other systems and applications

Where Interoperability is highly important: systems and applications like real-time decisioning, messaging, API-gateways and data streaming.

2. Co-existence: Ability to exist and function in parallel with other core transactional systems

Why Co-existence is paramount: to co-exist with other core transactional databases, data lakes and transient operational databases.

3. Manageability: Ability to deploy and manage lifecycle effectively on any infrastructure like on-premises or cloud

Where Manageability is a key factor: the need to deploy and manage entire life cycle on cloud or on-premises, ease of provisioning, ease monitoring and ease of gaining insights.

Figure 1:  The Three Tenets of Enterprise Frictionless Deployment

 

The Three Tenets of Enterprise Frictionless Deployment

As a strategic investment, we formed an Aerospike Ecosystems team to exclusively focus on developing solutions and add-on products that address the three crucial tenets indicated above, while our core server team continue to thrive delivering “unrivaled performance at scale.”

Real-Time Analysis Framework – An add-on to the Aerospike database

As enterprises look to transform digitally, they realize the growing importance of real-time business decisions. This requires them to process and analyze vast amounts of data during the processing time of their transaction windows as the stream of events flow through that window. The time to process and arrive at these decisions at this moment of business is less than a second in most cases. These business decisions map onto a wide spectrum spanning real-time, near real-time and even offline/batch. Yet, as the time for decisioning compresses, the need for combining stream event data with historical transaction data (in a fraction of a second) increases.

The challenge for businesses/enterprises to deploy these solutions is made even more complex when using disparate machine learning (ML) engines, artificial intelligence (AI) engines, and complex event processing (CEP) engines in the decisioning process. Enterprises either resort to operating on a small slice of transactional data to be analyzed along with stream event data or use memory intensive approach – neither of which is ideal.

Thus, we are pleased to announce the availability of the Aerospike Real-time Analysis Framework 1.0, the first add-on solution to the Aerospike Database, as a part of our ecosystems strategy. This solution addresses the need for combining stream event data with historical transaction data faster, better and cheaper. As the stream event data is flowing through streaming engines into the Real-time Analysis Framework, the Framework can join stream data with corresponding historical transaction data stored in the Aerospike database in real-time while allowing the analytics application developer to plug the combined data into ML, AI, and CEP decisioning engines.

The Aerospike Real-time Analysis Framework reference architecture is depicted below indicating how the add-on interoperates with Aerospike database and decisioning layers:

Figure 2: Aerospike Real-time Analysis Framework – reference architecture

 

Key features

  • Combines event stream data (from e.g. Spark streaming) with historical transaction data (stored in the Aerospike database) in real-time
  • Performs join and intersect operations to combine the two datasets
  • Spark MLLib engine plugin support (with a vision of extending the support for AI and CEP engines)
  • Spark interface support

Benefits

  • Lower TCO for real-time analytics by operating on larger datasets yet with a smaller cluster footprint
  • Gains closed-loop business insights by operating on both transactional and stream datasets
  • Rapidly develop using Spark libraries – no additional skill set required
  • Accelerate business insights by enabling decisions in seconds as opposed to hours or days

Prominent set of use cases where this add-on solution can rapidly add value to your organization include:

  • Real-time risk calculation in margin trading,
  • Real-time payment fraud detection
  • Personalization & recommendation engines
  • Industrial IoT

 

For more information, please feel free to review the Aerospike Real-time Analysis Framework solution brief.

About Author

mm

    Bharath Yadla, Vice President Product Strategy, Ecosystems for Aerospike

    All posts by this author
    Bharath is the Vice President Product Strategy, Ecosystems for Aerospike, focusing on Aerospike Connectors, Cloud, Clients & Tooling. Before joining Aerospike, he was with Persistent Systems, as Vice-President, Strategy & Solutions. Prior to that he was Associate Vice-President & Founding member of the Digital Transformation unit at HCL where he focused on driving their sales, solutions and strategy.