Open source database software and services specialist Percona has detailed what is now a full platform-play for the organisation’s IT stack. Ready to promise big things (in terms of scope, scale and software source-code suitability) Percona claims to have created a more unified experience for software developers and database administrators (DBAs) i.e. the two core halves of the DevOps equation, in order to monitor, manage, secure and optimise database environments.
The firm also stresses a new focus on supporting private DataBase-as-a-Service (DBaaS) functionality. The very logically-named Percona Platform spans the company’s expertise with databases including MySQL (a free relational database management system), PostgreSQL (a similarly free traditional relational database management system with similarities to MySQL) and MongoDB.
MongoDB is a no-schema document database, noSQL, JSON database with a free community version, a self-managed enterprise advanced option and a fully-hosted multi-cloud database service. As explained here by Brandon Chen, “PostgreSQL is an object-relational database, while MySQL is purely relational. This means PostgreSQL offers more complex data types and allows objects to inherit properties, but it also makes working with PostgreSQL more complex.
With all that database diversity out there – and with many customers potentially running all three of the databases here and more proprietary technology products alongside them – Percona says it is dedicated to bringing together what it calls services, management and automated insights into a single database management and insights product.
“Percona Platform includes AI-powered recommendations that help users improve performance, management and security, as well as making it easier to scale up and manage their instances over time,” said Donnie Berkholz, senior vice president of product at Percona.
What is an AI-powered database?
What’s an AI-powered recommendation for a database professional? It could be intelligence on a number of levels. One stream of database AI might look after areas such as providing alerts for when databases need to be patched with updates, provided with maintenance periods or even paused for larger system updates. A secondary application of AI for databases could involve intelligence designed to optimise system resources and drive computing (or storage, processing and analytics or other) power to a particular part of large-scale database system due to one or other particularly intense workload. A third area of AI-powered recommendation functionality might offer text-search capabilities, or perhaps Machine Learning (ML) tools to get smart on tasks like data ingestion, data exploration, data analytics itself and subsequent data visualisation. A fourth cornerstone of database AI might include all of the above and some other newer prototyped approaches to using analytics (typically involving more user-level self-service functions) as yet unknown.
In light of all these concepts and technology delivery points then, Berkholz asserts that, with Percona Platform, the company thought about how to improve the experience for developers and DBAs around running their data workloads and how to help people achieve their goals faster around managing databases as part of their overall approach.
“Developers want to improve their efficiency and experience across their tasks and databases are no exception. Improving this will be an area of investment for companies – whether this involves time for their staff to implement projects, or investing in cloud services to provide the tools they need,” said Bradley Shimmin, chief analyst for AI platforms, analytics and data management at Omdia.
An explosion of enterprise data tools
Shimmin suggests that Percona’s approach to this challenge – which is markedly and specifically aligned to open source technologies – aims to provide self-managed control and internal ownership as well as the flexibility and ease of use that many companies tout with their DBaaS offerings. All of this is happening in the face of what other industry analysts such as RedMonk’s Rachel Stephens calls an, ‘explosion of enterprise data tools’, which has led to rampant fragmentation and highly complex, disjointed workflows.
“Organizations that prize velocity are looking for ways to integrate various data tools into their broader development workflows and monitor them. This is the opportunity that Percona Platform is built for. The combination of software and services in this space is intended to help customers better manage their database deployments,” commented RedMonk’s Stephens.
Percona has now woven over a decade of open source database expertise into what company president Ann Schlemmer says is now a consistent, modern database experience for all organizations, from any cloud to on-premises.
Talking about how Percona has changed now that the company’s technology has progress to full-blown platform-level offering, Schlemmer points to automated insights, self-service operations and some of the best database observability and management tools in the market.
“This launch will help us scale our business and create a sustainable model in the age of cloud, while respecting the freedom and accessibility that is open source. We have seen other companies turn away from open source and adopt closed source licenses in order to maximize their profitability, which we think is not in the best interests of customers. Our approach makes our expertise available to all our customers and users, putting their needs first,” added Schlemmer.
Is Percona really a platform?
Taking stock of all this then, can we say that Percona has truly elevated itself to platform level status? As we have discussed here before, the move to platform inherently implies that other people, products and processes can build ‘on top of’ your base platform technology and use it as a foundational substrate underlay.
The use of the term platform in this sense is arguably slightly evolved from the way we might talk about Google Maps, Facebook or even Microsoft Office or 365 as a platform. In those cases, we can build new software services that draw from, connect to and integrate with the base layer in order to create something new – but that new technology normally carries with it some visually detectable element of the base layer DNA in its Graphical User Interface.
In the case of Percona Platform, we can still build new software services that draw from, connect to and integrate with the base layer in order to create something new – but that new technology is shaped by a less visible (or wholly invisible) lower automation, intelligence and efficiency layer delivered by virtue of the smart database management going on underneath.
How lower database platform functions work
How that lower database platform functionality manifests and works is the crux of this matter.
With Percona Platform, software application developers are able to make use of what is known as a Private DataBase-as-a-Service (DBaaS), that allows them to quickly provision and management their database requirements on a self-service basis across both the test and the live production instance phase of their work.
Software developers looking to ‘platform off’ Percona Platform also get Query Analytics functions. This enables them to get visibility into the working core of the database being used to support the applications they build. In working practice, this means being able to find performance patterns and understand what they mean in the context of diagnostics and troubleshooting.
Percona has also incorporated a range of what it calls ‘advisors’ to provide automated insights that ensure user databases are performing well n terms of security management, configuration and performance.
At an absolutely purist (and deeper) technologist level, Percona is talking about its platform in the context of these data management services at the core, then a layer of data (in a database itself), before then also layering on a runtime layer where (in basic terms) the application itself all comes to life. Sit all that on Amazon Web Services (AWS), Google Cloud Platform (GCP), Microsoft Azure, or some other non-big-3 cloud, or on-premises… and you have a construct architecture of data tools running in what you can only call a platform.
Applications and data-centric services are becoming increasingly complex, often fragmented and increasingly connected, creating database services that work as a platform in and of themselves is what is happening here.
Just like a surfboard, you may not always know what’s beneath you, but at the end of the day, you do know you’re standing on a platform.