Raising the Curtain on Powerful, Flexible, and Scalable Applications

Behind every successful, modern application is a variety of technologies and strategies that invite greater flexibility, efficiency, and innovation. While ideal, this is easier said than done; many “traditional” approaches to data management—still in place at many enterprises—are ill-equipped to power a truly distributed, modular, and portable application.

Experts joined DBTA’s webinar, Powering Modern Applications: Data Management for Speed, Scale, and Flexibility, to discuss key solutions and emerging best practices that properly underpin successful applications that are fast, agile, and easy to use.

Matthew Groves, DevRel engineer at Couchbase, explained that application needs have changed. Customers of Couchbase have expressed that they want to deliver great experiences, develop those experiences efficiently, and deploy effectively, to multiple places and devices.

These emerging needs are further compounded by the changing desires for databases, where databases are now expected to handle massive numbers of users, scale and perform, leverage microservices, and deliver web, mobile, and IoT experiences, said Groves.

Particularly in the realm of speed, scale, and flexibility, Couchbase Capella delivers sub-millisecond response time for data operations, active-active clustering for more efficient scaling, and JSON schema flexibility for greater agility, paired with a multi-model approach.

Couchbase Capella addresses modern data sprawl and management challenges. Michael O'Donnell (PhD), senior analyst at Quest, explained that most enterprises are bogged down by a “big ball of mud,” or “monolithic system architectures [that are] difficult to understand, hard to maintain, and tightly coupled because it has many dependencies.” This big ball of mud ultimately inhibits data management due to its unstructured, hindering speed, scalability, and flexibility.

To remediate this phenomenon, Quest offers solutions for both data management and data intelligence that unite modern data architectures often wrought with tool sprawl and an over-proliferation of solutions. Quest boils down the power of its erwin by Quest data modeling and intelligence solution into seven steps that the solution follows to maximize data value:

  1. Model: Design data architecture
  2. Catalog: Search and find data easily
  3. Curate: Enrich with business context
  4. Govern: Apply business rules and policies
  5. Observe: Raise data visibility to proactively manage
  6. Score: Automate data profiling and quality scoring
  7. Shop: Make trusted, governed data widely accessible

Dan DeMers, CEO of Cinchy, stated it plainly: Data integration is awesome, but it also sucks. While integration is crucial to centralizing data in an organization, it simultaneously wastes resources and becomes more expensive as more technologies and digital transformation grows.

DeMers further explained that we’ve seen similar trends with software; like software (or code), data needs to be federated (i.e., collaborative) to scale. 

“The reality is that code belongs as a network on one plane, and data, similarly, belongs as a network but on a separate and distinct plane,” DeMers said. “That is the inevitable future, where it’s an interconnected, highly collaborative, highly controlled network. This is what unlocks scale and frees IT organizations to be able to deliver business value instead of endlessly moving data around in an organization.”

Ultimately, collaboration replaces integration when there are multiple producers. Simply sharing is inadequate; co-production is the reality—and the future of data—that must be accommodated, according to DeMers. Applying collaboration helps to alleviate the pressures of data integration, all while enabling data owners to maintain control.

Anil Inamdar, director of consulting at Instaclustr, now part of Spot by Netapp, focused on open source that is really open source, bulldozing the barriers that often limit its implementation.

Inamdar highlighted five key considerations to address—and overcome—when it comes to open source adoption: licensing, governance, business models, and ecosystem.

While free, open source licenses are perpetual entities that enable the copyright holder to provide permission for other people to use their code. Licenses are often complicated and complex, spanning three broad categories: permissive licenses, copyleft licenses, and custom licenses.

Regarding governance, this area concerns those who make critical decisions on the code base. Inamdar explained that there are, broadly, two types: non-profit foundations, which provide governance and decision making, and for-profit corporations, which retain a unilateral decision-making ability as to what changes to accept, as well as if they want to change the license in future versions.

Inamdar further urged viewers to take business models into account when deciding on an open source software, as solutions can range from purely open source to open core, open code or open source IP owner, and cloud providers.

Inamdar introduced Instaclustr, a 100% open source database solutions provider, supporting technologies like Cassandra, Kafka, Redis, Spark, and more. Its managed platform offers 24/7 support, certified and tested builds, and full lifecycle management, filling the gaps for open source ecosystems.

For an in-depth review of technologies and strategies for powering modern applications, featuring use cases, a Q&A, and more, you can watch an archived version of the webinar here.