Apps

How to push PaaS usage beyond 12-factor apps

Comment

Software, web development, programming concept. Abstract Programming language and program code on screen laptop. Laptop and icons company network . Technology process of Software development
Image Credits: Andrey Suslov / Getty Images

Sylvain Kalache

Contributor

Sylvain Kalache is the co-founder of Holberton, an edtech company training digital talent in more than 10 countries. An entrepreneur and software engineer, he has worked in the tech industry for more than a decade. Part of the team that led SlideShare to be acquired by LinkedIn, he has written for CIO and VentureBeat.

More posts from Sylvain Kalache

Platform-as-a-service (PaaS) emerged as a leading force in the ever-evolving quest to streamline software development. PaaS dates back to 2006 with Force.com, followed by Heroku, AWS Elastic Beanstalk, and DotCloud, which later transformed into Docker.

While the PaaS sector commands a substantial $170 billion market share within the cloud industry, companies still grapple with manual deployment and workload life cycle management today. So why isn’t platform-as-a-service more widely adopted?

Providing a PaaS experience across all workloads

PaaS platforms could be more versatile, and I am not speaking of language and framework compatibility. While PaaS is often defined as a one-stop shop for deploying any application, there is a catch. By applications, what is usually implied here is 12-factor applications.

However, many workloads don’t neatly fit the mold of typical web apps; they come with unique requirements, such as batch processing jobs, high-performance computing (HPC) workloads, GPU-intensive tasks, data-centric applications, or even quantum computing workloads.

I won’t go over all the advantages that PaaS provides. Still, companies should manage all their workloads in the easiest way possible, and abstracting their deployment and management is the way to go.

A shift is needed. First, companies embracing the PaaS paradigm must recognize that there won’t be a one-size-fits-all workload solution. In a recent discourse on the topic, former Google engineer Kelsey Hightower reinforces this notion that a single, all-encompassing PaaS remains improbable.

He also used workload API to designate a tool that provides this seamless “here is my app, run it for me” experience. I like the term “workload API” because it clearly states the intent: to run a specific workload. Compared to platform-as-a-service (PaaS), which needs to be more accurate and leads to this confusion that PaaS is a silver bullet to running anything. I will use this term for the remainder of the article.

The second change for companies wanting to provide a seamless deployment and management experience for all their workloads is considering that each workload should have its workload API. For example, Amazon Lambda could be used for batch jobs, Vercel for front end, Vertex AI for machine learning models, and Korifi for web apps.

Now, let’s explore how to pick workload APIs.

Managing the lock-in concern

While vendor lock-in can be a valid worry, it is more manageable than it appears.

Look for platforms that are following standards. For example, many providers built their services on a shared foundation. For instance, BOSH, SAP Business Technology Platform, and VMware Tanzu are all based on open source Cloud Foundry, which makes transitioning between these platforms more manageable.

Another example is picking a tool that works with containers, which act as standardized building blocks — making it easier to migrate between workload APIs that support containers without encountering significant roadblocks.

Look for workload APIs that are GitOps-based or at least compatible. Because any codebase, no matter the workload type, generally lives on a Git repository. Examples of GitOps platforms include Weaveworks and Kubefirst.

For machine learning models, look for workload APIs that support common ML frameworks like open source TensorFlow, PyTorch, or scikit-learn. Also, make sure they accept open source formats like ONNX or PMML. Finally, another bonus is compatibility with platforms designed to help manage the machine learning life cycle, such as open source MLflow and Kubeflow.

While I cannot review every possible workload, integration with open source solutions and open standards is the common thread.

Finally, if you still need to decide between a few options, look at the exit routes. Workload API providers want your business and often build migration tools to import your workloads from another platform. Make sure to consider this in your selection process.

Making it cost-effective

While workload API provider costs constantly decrease, they can be higher than running your servers. However, consider the equation with a TOC (total cost of ownership) in mind.

As mentioned above, paying for a workload API service will likely be more expensive than running your infrastructure. Still, you will also need to hire a team large enough to build, maintain, improve, and secure your infrastructure. Also, companies specializing in workload APIs will provide a better solution with less downtime, better scalability, and performance, leading you to more cost reductions down the road.

For some workloads, such as machine learning or HPC (high-performance computing), the cost of buying the hardware can be prohibitive, or the hardware might not be available. For example, with the rise of LLMs, there has been a strong demand for machines that can run GPU-based workloads, leading to shortages. On the other hand, workload API providers may own or book up the computing capacity they need to serve their customers, meaning higher availability.

It would be best if you also baked in the potential revenue impact of opportunity loss. Your competition might be using workload APIs, increasing the velocity at which they can test and deploy new features and provide a smoother customer experience, ultimately driving more business to their product.

Finally, using open source options — such as Dokku and Korifi for web app workloads — is an alternative solution. They allow you to control the infrastructure and operation costs without paying for a software license. Large organizations with specific needs can also modify the code as they see fit or use the software as building blocks.

Understanding what is under the hood

While abstracting the complexity from end users is essential, understanding the inner workings of your chosen workload API is equally crucial. Ensure it aligns with all your technical requirements, spanning scalability, security, and reliability. A deep understanding will also allow you to troubleshoot effectively when things go awry.

And because workload APIs are, by nature, straightforward to set up and use, do not hesitate to try them out to verify if they can accommodate the workload you are trying to migrate or if they fit the requirements you have in mind. It will often be faster than going through the marketing material and documentation.

Watch out for data compliance

With 70% of countries having data compliance legislation in place, this is not a topic that can be ignored. If you are not hosting your PaaS, ensure that the provider data compliance and security standards match the regulations you must abide by. While every jurisdiction will have its requirements, looking for data security and encryption, residency, privacy policies, and retention is a good start.

Once you have multiple workload APIs in place, you can take control of some of the underlying parts — for example, by putting an IaC between your workload APIs and the public cloud to provision resources. This way, you can ensure that any security or compliance requirements are met across your workloads. Tools such as Terraform, OpenTofu, and Pulumi are popular choices.

Come up with a strategy

Once you’ve picked the tools, migrate progressively. Start with a noncritical workload to mitigate the impact of potential instability. Learn from these initial experiences before advancing to mission-critical workloads.

Once you’ve deployed a workload API, you must gauge user satisfaction: developer experience. A tool is as good as its adoption rate. If practitioners don’t like it, they may stop using it and find some alternative. This is especially true for large organizations.

Finally, ensure that the migrated workload has been through multiple life cycles to provide comprehensive coverage. For instance, verify if the workload API can handle sudden load spikes and that a backup and restoration operation ran successfully. This progressive approach will ensure that you learn from potential failures and keep them in mind when you migrate your other workloads.

The effort is worth it

Moving to workload APIs is worth it. Cost savings, faster development, better scalability, reduced downtime, and improved productivity are benefits that organizations are experiencing. It’s time to apply this concept to all your organization’s workloads.

More TechCrunch

The prospects for troubled banking-as-a-service startup Synapse have gone from bad to worse this week after a United States Trustee filed an emergency motion on Wednesday.  The trustee is asking…

A US Trustee wants troubled fintech Synapse to be liquidated via Chapter 7 bankruptcy, cites ‘gross mismanagement’

U.K.-based Seraphim Space is spinning up its 13th accelerator program, with nine participating companies working on a range of tech from propulsion to in-space manufacturing and space situational awareness. The…

Seraphim’s latest space accelerator welcomes nine companies

OpenAI has reached a deal with Reddit to use the social news site’s data for training AI models. In a blog post on OpenAI’s press relations site, the company said…

OpenAI inks deal to train AI on Reddit data

X users will now be able to discover posts from new Communities that are trending directly from an Explore tab within the section.

X pushes more users to Communities

For Mark Zuckerberg’s 40th birthday, his wife got him a photoshoot. Zuckerberg gives the camera a sly smile as he sits amid a carefully crafted re-creation of his childhood bedroom.…

Mark Zuckerberg’s makeover: Midlife crisis or carefully crafted rebrand?

Strava announced a slew of features, including AI to weed out leaderboard cheats, a new ‘family’ subscription plan, dark mode and more.

Strava taps AI to weed out leaderboard cheats, unveils ‘family’ plan, dark mode and more

We all fall down sometimes. Astronauts are no exception. You need to be in peak physical condition for space travel, but bulky space suits and lower gravity levels can be…

Astronauts fall over. Robotic limbs can help them back up.

Microsoft will launch its custom Cobalt 100 chips to customers as a public preview at its Build conference next week, TechCrunch has learned. In an analyst briefing ahead of Build,…

Microsoft’s custom Cobalt chips will come to Azure next week

What a wild week for transportation news! It was a smorgasbord of news that seemed to touch every sector and theme in transportation.

Tesla keeps cutting jobs and the feds probe Waymo

Sony Music Group has sent letters to more than 700 tech companies and music streaming services to warn them not to use its music to train AI without explicit permission.…

Sony Music warns tech companies over ‘unauthorized’ use of its content to train AI

Winston Chi, Butter’s founder and CEO, told TechCrunch that “most parties, including our investors and us, are making money” from the exit.

GrubMarket buys Butter to give its food distribution tech an AI boost

The investor lawsuit is related to Bolt securing a $30 million personal loan to Ryan Breslow, which was later defaulted on.

Bolt founder Ryan Breslow wants to settle an investor lawsuit by returning $37 million worth of shares

Meta, the parent company of Facebook, launched an enterprise version of the prominent social network in 2015. It always seemed like a stretch for a company built on a consumer…

With the end of Workplace, it’s fair to wonder if Meta was ever serious about the enterprise

X, formerly Twitter, turned TweetDeck into X Pro and pushed it behind a paywall. But there is a new column-based social media tool in town, and it’s from Instagram Threads.…

Meta Threads is testing pinned columns on the web, similar to the old TweetDeck

As part of 2024’s Accessibility Awareness Day, Google is showing off some updates to Android that should be useful to folks with mobility or vision impairments. Project Gameface allows gamers…

Google expands hands-free and eyes-free interfaces on Android

A hacker listed the data allegedly breached from Samco on a known cybercrime forum.

Hacker claims theft of India’s Samco account data

A top European privacy watchdog is investigating following the recent breaches of Dell customers’ personal information, TechCrunch has learned.  Ireland’s Data Protection Commission (DPC) deputy commissioner Graham Doyle confirmed to…

Ireland privacy watchdog confirms Dell data breach investigation

Ampere and Qualcomm aren’t the most obvious of partners. Both, after all, offer Arm-based chips for running data center servers (though Qualcomm’s largest market remains mobile). But as the two…

Ampere teams up with Qualcomm to launch an Arm-based AI server

At Google’s I/O developer conference, the company made its case to developers — and to some extent, consumers — why its bets on AI are ahead of rivals. At the…

Google I/O was an AI evolution, not a revolution

TechCrunch Disrupt has always been the ultimate convergence point for all things startup and tech. In the bustling world of innovation, it serves as the “big top” tent, where entrepreneurs,…

Meet the Magnificent Six: A tour of the stages at Disrupt 2024

There’s apparently a lot of demand for an on-demand handyperson. Khosla Ventures and Pear VC have just tripled down on their investment in Honey Homes, which offers up a dedicated…

Khosla Ventures, Pear VC triple down on Honey Homes, a smart way to hire a handyman

TikTok is testing the ability for users to upload 60-minute videos, the company confirmed to TechCrunch on Thursday. The feature is available to a limited group of users in select…

TikTok tests 60-minute video uploads as it continues to take on YouTube

Flock Safety is a multibillion-dollar startup that’s got eyes everywhere. As of Wednesday, with the company’s new Solar Condor cameras, those eyes are solar-powered and use wireless 5G networks to…

Flock Safety’s solar-powered cameras could make surveillance more widespread

Since he was very young, Bar Mor knew that he would inevitably do something with real estate. His family was involved in all types of real estate projects, from ground-up…

Agora raises $34M Series B to keep building the Carta for real estate

Poshmark, the social commerce site that lets people buy and sell new and used items to each other, launched a paid marketing tool on Thursday, giving sellers the ability to…

Poshmark’s ‘Promoted Closet’ tool lets sellers boost all their listings at once

Google is launching a Gemini add-on for educational institutes through Google Workspace.

Google adds Gemini to its Education suite

More money for the generative AI boom: Y Combinator-backed developer infrastructure startup Recall.ai announced Thursday it has raised a $10 million Series A funding round, bringing its total raised to over…

YC-backed Recall.ai gets $10M Series A to help companies use virtual meeting data

Engineers Adam Keating and Jeremy Andrews were tired of using spreadsheets and screenshots to collab with teammates — so they launched a startup, CoLab, to build a better way. The…

CoLab’s collaborative tools for engineers line up $21M in new funding

Reddit announced on Wednesday that it is reintroducing its awards system after shutting down the program last year. The company said that most of the mechanisms related to awards will…

Reddit reintroduces its awards system

Sigma Computing, a startup building a range of data analytics and business intelligence tools, has raised $200 million in a fresh VC round.

Sigma is building a suite of collaborative data analytics tools