Anurag Goel, Render founder and CEO, told VentureBeat via email that the company’s core revolves around automating tasks that typically require large teams of devops engineers working on cloud providers like AWS and GCP. Goel said these include routine tasks like pulling code from GitHub, building, deploying, monitoring in real-time, and scaling them on demand.
“With it, we’re creating an entirely new abstraction for cloud infrastructure: with something like AWS, you can spin up a virtual machine with an API call, but you still have to install application dependencies, configure security groups, IAM, SSL, security updates, and more. In contrast, Render customers can spin up full-fledged services without managing servers or virtual machines,” the company said in a press release.
Goel explained that Render’s platform is a general-purpose compute platform that allows its customers to deploy complex applications and data stacks without involving a devops team. This enables enterprise data science and engineering teams to move more quickly and focus on the actual data instead of the underlying infrastructure.
Advancing cloud infrastructure
Render claims it’s advancing cloud infrastructure in an easy-to-use manner — enabling customers to build platforms by clicking a button without any devops expertise. The company explained that some examples of what its platform enables its customers to do includes HTTP/3, private networking, persistent disk storage, native CPU and memory-based autoscaling, zero-downtime deploys, industry-leading DDoS protection, automatic staging and development environments, and infrastructure-as-code.
According to Goel, Render is the first platform-as-a-service to offer built-in DDoS protection. As DDoS attacks continue to rise in complexity, volume and frequency, they threaten the network security of even the smallest enterprises, according to Gartner. Building DDoS protection is typically time-consuming and an expensive endeavor that requires in-depth expertise few engineers have. Goel noted that this is why DDoS protection is prioritized t of every Render-hosted service for free.
Two of these major services include the following:
- Redis: A cache, a queue, a subserver, a real-time stream processor, and a database all in one. However, managing and scaling Redis in production — Goel said— is time-consuming, complex, and precisely the kind of undifferentiated devops work Render automates. With Render-managed Redis, the company claims developers can focus on using Redis effectively without worrying about reliability or scalability.
- SSH: Although this is the most well-known mechanism to connect to remote machines securely, Goel said managing SSH servers is complex. SSH server management involves manually copying SSH keys, keeping them up to date, configuring firewalls and ports, and protecting against abuse. Render says its Render SSH is fully managed and needs no administration — automatically managing keys, security, and connections. The company says Render SSH also empowers developers to use a familiar interface to connect to their cloud applications instead of cloud-specific command line interfaces (CLI)s.
Improving cloud hosting accessibility
Render claims it’s making cloud hosting more accessible and powerful with every new feature.
“Redis, SSH, and DDoS protection are critical components of most software stacks, and our goal with core features like these is to help its existing customers grow on Render without constraints. At the same time, we’ve spoken to hundreds of existing Heroku customers who want to move to Render and need features like Redis or SSH to make the switch. We expect many longtime Heroku users will be able to make the switch with this launch,” the company explained in a statement.
Render’s new public API will empower a new generation of content management systems, site builders, and open source projects to offer cloud hosting to their customers, according to Goel.
“For example, instead of building out a devops team to manage infrastructure on AWS, the next big e-commerce company can use Render’s API to create customer storefronts in just a few API calls. Similarly, many site builders like Help Site already use Render’s API under the hood to offer fully managed hosting to their customers,” he added.
He also said that rapidly growing open source projects like Daxter and Strapi will now be able to use Render to build (or supplement) their managed cloud offerings — all without needing an army of devops engineers.
Overall players in the space
While Render claims that its automation, performance, and flexibility differentiates it from other players in the industry. The players in the market fall into two categories, said Goel: infrastructure-as-a-service (IaaS) providers like the big three clouds AWS, Azure, and Google Cloud Platform; and platform-as-a-service (PaaS) providers like Heroku.
Render claims it differentiates itself from other IaaS providers because it automates devops and focuses on startups and mid-market companies that don’t already have devops engineers. However, Goel said, IaaS platforms force customers to do repetitive and error-prone devops work and target enterprise organizations with massive devops teams.
Furthermore, the company says it differentiates itself from existing PaaS providers by offering more performance and flexibility at a competitive price.
“As a simple example, Heroku doesn’t support workloads that need more than 14GB in memory, whereas workloads on Render can consume over 512GB. We continue to build features to help companies grow on Render and stay on the platform without being forced to migrate to IaaS providers for functionality and cost reasons,” Goel said.
Render’s AI and ML capabilities
Render says its customers run complex ML and data processing workloads in a few primary ways:
- Creating one-off data processing jobs using our public API.
- Using PostgreSQL read replicas to run expensive queries that can’t run on production databases.
- Using its built-in CPU and memory-based auto scaling to launch hundreds of parallel background workers for distributed computing.
The company says it plans to make it even easier to host AI and ML workloads on Render by introducing compute-optimized and GPU instances for both training and inference workloads. The company cited Goel’s experience with building Crestle, the first serverless Jupyter notebook provider backed by GPU instances, as a bedrock for this plan.
Previous and upcoming milestones to expect
Render announced $20 million in new funding as part of its Series A round in late November 2021. The company says it will launch new hosting regions, especially in APAC, and more products and features to help its Heroku customers migrate to Render even faster.
Render’s employee headcount is approaching 40 and the company expects to double this number over the next year. A few of Render’s customers include startups like Clearbit, Mux, NEAR, and public companies like Anker.