Once upon an IT time, everything was a “point product,” a specific application designed to do a single job inside a desktop PC, server, storage array, network, or mobile device. Point solutions are still used every day in many enterprise systems, but as IT continues to evolve, the platform approach beats point solutions in almost […]
Once upon an IT time, everything was a “point product,” a specific application designed to do a single job inside a desktop PC, server, storage array, network, or mobile device. Point solutions are still used every day in many enterprise systems, but as IT continues to evolve, the platform approach beats point solutions in almost every use case.
A few years ago, there were several choices of data deduplication apps for storage, and now, it’s a standard function in every system. Different vendors offered security options (there was an anti-virus app, a home network security app, and endpoint protection software); now, it all comes bundled as one solution in each vendor’s package. Similar stories occurred in other IT sectors.
Unlike point products that solve one specific business problem, platforms are software packages that enable users to organize functions into one console and also design new solutions as use cases evolve. Platforms help businesses to develop end-to-end solutions that work for their specific needs and change with them. For one example, Cloudera’s enterprise data cloud is a platform designed specifically for improving control, connectivity, and data flow inside a data center in addition to public, private and hybrid clouds. It incorporates all the key functions–which used to consist of a dozen or more point-product apps–of running an enterprise hybrid cloud system into a centralized control console. Having only one vendor to call for updates, security patches, questions–or in the event of a problem–is a major plus for administrators.
Business executives are some of the biggest platform supporters. CIOs and CFOs have regular headaches about handling all those point-product vendors and their legalese, rules, and regulations — not to mention limitations of how each product plays in a proprietary system with lots of other software and hardware that may not align well.
Consider the way cloud computing worked as a point solution: Back in the mid-to-late 1990s, the earliest type of what we now know as a “cloud service” was called an ASP (application service provider). These were single-point applications delivered via the internet from one data center to another — all private connections — and while this was pioneering tech, everything about them was brittle and unreliable.
Bandwidth was limited; heavy data loads took days or weeks to move from Point A to Point B. There was only password-based security and mostly hit-or-miss backup for data flows, and if one connecting network handshake was misconfigured, data loads blew up.
Now we’re immersed in a full-fledged IT upgrade trend in which users subscribe to one cloud provider for a platform of individual services that they can use or not use, depending on their use-case needs. These services are provided on a menu for users to select, and most often, they pay “by the drink” for each service.
Companies generally have hundreds or thousands of applications and only a few platform providers, and business users love this. They need to monitor only one monthly bill per platform, rather than per each application.
Platform companies team up on a regular basis, which is another benefit for enterprise users. Because they build products with collaboration in mind, platform makers know how to align their products with other companies inside the context of an enterprise system. With so much mix-and-matching being done by CTOs as requirements for use cases shift — especially in cloud services — companies such as Cloudera have made their technology connectible as needed for their customers.
For example, Cloudera Data Platform provides a big-data platform designed for deployment in any type of enterprise system–physical or in a cloud. Companies select this because it provides a central, scalable, and secure environment for handling workloads for batch, interactive, and real-time analytics.
Because CDP’s menu includes a dedicated database, storage warehouse, multifunction analytics, governance, and well-supported environments for engineering and other third-party services, all the key components are available to run an enterprise cloud either in-house or in a hybrid situation, using clouds that include AWS, GCP and Azure.
But the most important benefit here is performance.
Why do point products generally come up short in delivering what full solution platforms provide? Certainly, there is value for point products in specific use cases, but those are generally legacy systems designed for point products, and legacy systems — even if they work just fine — won’t be around forever.
This is even more evident with big data. SaaS platforms are nearly always data lifecycle products, said Patrick Moorhead, president and principal analyst at Moor Insights & Strategy.
“The benefits of using an end-to-end data platform,” Moorhead said, “are that fewer third-party handoffs mean higher accuracy of data going where it should be, and automated so it takes less time.
“Data is more secure and it is easier to pass a data audit by a regulator, because the data is governed under the same set of rules every step of the way.”
Charles King, IT analyst with Pund-IT, said that the value of SaaS and on-prem solutions reflect larger trends in hybrid cloud systems that have effectively become the IT platform of choice for modern businesses.
“Partly that’s due to the relative ease they offer in terms of both initial sizing and scaling up to meet the needs of growing organizations,” King said. “You can see it in the way that enterprise vendors are updating and enhancing the flexibility of both hardware and software offerings to accommodate hybrid cloud solutions and strategies.
Point products aren’t likely to disappear entirely, King said. “They’ll always have a place in specialized infrastructures and environments that enterprises maintain on site to meet specific security or service-level requirements. But for most enterprise applications and workloads, supporting SaaS and on-prem separately or in hybrid cloud environments represents both current thinking and the future.”
Hybrid clouds and data platforms offer one of the most straightforward routes for forward-looking enterprise leaders. What’s more, the simplicity and convenience of platforms entice not only business executives but IT as well.
“IT is generally overworked and underfunded,” explained Rob Enderle, former software developer and principal analyst at Enderle Group. “Thus it tends to prefer product bundles made up of integrated and consistently managed offerings so that their already nearly unmanageable complexity isn’t made worse.
“As a result, IT generally will prefer an integrated offering with a needed function or feature that isn’t as good operationally as a point product that does the same thing, because the complexity problem it introduces overcomes the operational advantages it provides. IT will often work harder to implement products that reduce complexity while avoiding point offerings that increase it.”
The platform vendor sees faster growth and can quickly ramp resources to support the customers’ needs, Jon Toor, a 30-year industry veteran and an executive at data storage provider Cloudian (a partner of Cloudera), said. “This virtuous cycle benefits both buyers and sellers, as long as the platform succeeds in attracting vendors,” he said. “Hence Steve Ballmer’s well-known rant: ‘Developers, developers, developers!’”
The virtuous cycle did not work for OpenStack, Toor said. “With many (open source) masters, the OpenStack platform failed to achieve simplicity. Vendors were interested in OpenStack, but customers saw better, more accessible platforms.”
Competitors were providing a user interface that was much easier to navigate, being designed for a general audience — unlike OpenStack that was aimed specifically at developers.
OpenStack is a lesson in the value of an intuitive, easy-to-use user interface and concise, clear documentation, even in platform solutions for professionals. The late Steve Jobs once said that no matter how good the software may be, it’s useless if nobody can figure out how to use it.
Toor points to a famous example of platform success: VMware. “It offers a powerful platform that has solved problems in virtually every industry,” Toor said. “And because it was easy to use, IT people could deliver value right away, fueling rapid customer growth and further expansion of the ecosystem. The virtuous cycle worked.”
The user interface and VMware’s vast set of templates and reference architectures were a huge advantage, because IT staff could move quickly in getting tasks done, instead of spending time learning how to use the platform.
Business in next-gen IT is getting done here in 2021 when line-of-business employees at an enterprise can simply log in to the platform, input their instructions, and go to work. Collaboration, simplicity, and automation are paths to success, and a platform approach is the vehicle.
So why do platforms beat point solutions in every business case? When you have choices, as platforms offer, there are many potential avenues to a successful outcome of an IT project. Using a point product, there is only one lonely road to take, and it might not be the best one.
Read here the successes Cloudera customers have had with Cloudera Data Platform (CDP).
The post Five Reasons Why Platforms Beat Point Solutions in Every Business Case appeared first on Cloudera Blog.