By Tod Golding, Principal Partner Solutions Architect, AWS SaaS Factory
By Ashish Kathapurkar, Sr. Partner Manager, AWS SaaS Factory
The AWS SaaS Factory team recently had the chance to work with Ridgeline, a privately funded fintech startup determined to modernize investment management software, and help build their purpose-built enterprise platform on Amazon Web Services (AWS).
Ridgeline chose AWS as its cloud platform for the breadth and depth of offerings. Using AWS services, Ridgeline was able to build a fault-tolerant solution that scales on demand and adheres to compliance standards such as System and Organizational Controls (SOC) Type II.
Ridgeline founder and CEO Dave Duffield has spent the last four decades building and growing value-based enterprise software companies. PeopleSoft, which Duffield founded and led, was a pioneer of client/server technology. Workday, his subsequent venture, was one of the first enterprise platforms born on the cloud. Ridgeline is taking the next step with AWS to build an industry-specific, enterprise platform on the public cloud.
The AWS SaaS Factory team sat down with Chris Petrunik, Director of Engineering at Ridgeline, to learn what they’re launching and how it benefits customers. We also asked what advice they have for other businesses building software-as-a-service (SaaS) solutions on AWS.
To accelerate and guide the adoption of a SaaS delivery model on AWS, we launched AWS SaaS Factory in 2018. This program offers a comprehensive set of business and technical best practices, as well as engagement opportunities that empower and support independent software vendors (ISVs) building SaaS offerings on AWS.
Q&A with Ridgeline
SaaS Factory: Can you tell us about your background and personal experience with cloud computing?
Chris Petrunik: I joined Ridgeline [in 2019] to help build our engineering organization. While my current work is focused on both DevOps and UI engineering, my background is in application development. I spent five years at Workday prior to joining Ridgeline; there I worked with a great team, Workday Student, to bring higher education into the cloud.
Similar to Ridgeline’s customers in the financial sector, universities across the world rely on mainframe-based systems that are often decades old. Prior to Workday, I spent time building risk management systems at JP Morgan. I am excited to help bring modern technology to the industry where I started my career.
SaaS Factory: What products and solutions has Ridgeline built on AWS? How does the Ridgeline platform work?
Chris Petrunik: Ridgeline is building a software platform for investment management firms—one system to replace a disconnected suite of applications. Our mission requires reimagining core infrastructure on modern, cloud technology. Our product will be a true “enterprise” platform, which means one version of data and positions, one version of the software, one security model, and one book of record. It also means designing a robust, human-readable API for getting data into and out of the system.
Internally, we organize our platform into five functional areas: data, accounting, trading, reporting, and client experience. Ridgeline is data-agnostic and processes feeds from a variety of sources. Our accounting engine is double-entry, multi-asset, multi-currency, and provides granular detail for analytics.
Our trading system allows the middle office to construct portfolios and easily comply with restrictions. The reporting team is focused on surfacing internal data to portfolio managers and richly-detailed reports to investors. Finally, the client experience system allows firms to provide world-class service quickly and easily.
Our entire platform is built using the AWS Cloud—we’re all in on AWS!
SaaS Factory: What’s the opportunity for customers? How does Ridgeline address that opportunity?
Chris Petrunik: Investment managers are experiencing a period of rapid transition. Money is increasingly moving from active to passive vehicles. The regulatory environment is changing. Profit margins are shrinking. Meanwhile, software upgrades are painful and expensive. Integration headaches are common. Too many people are doing too much manual work. Our belief is that firms have been underserved by legacy technology.
Specifically, investment managers are gridlocked by data silos, outdated infrastructure, and legacy software investments. It’s a tough combination of forces—industry trends are requiring firms to become more efficient, while their software tools are making this difficult.
At Ridgeline, we’ve dedicated ourselves to rebuilding infrastructure and software from a clean sheet of paper. Our goal is to help firms provide differentiated service, reduce risk, and outmaneuver the competition
SaaS Factory: Can you walk us through the architecture? What AWS services are key?
Chris Petrunik: Ridgeline’s journey started with the decision to use cloud-native technologies, and we are built entirely on AWS. We’ve also made a bet that starting with a serverless architecture will be key to our long-term success. A major benefit of this technology is how it enforces more modular software design.
Decomposing a traditional monolith—or even a traditional microservice—into small functions often requires a complete rewrite to fully take advantage of serverless benefits. It’s one reason why you won’t see many existing applications transition to serverless. By embracing a serverless architecture from the start, our software is highly-modular and therefore highly-evolvable. Each function can be tested independently and rewritten easily should the need arise.
At the storage and persistence layer, Ridgeline leverages Amazon Simple Storage Service (Amazon S3) for file storage, Amazon DynamoDB for key/value-based datasets, and Amazon Aurora for more traditional relational data. For compute, we utilize AWS Lambda with AWS Step Functions layered on top for larger processes.
While we use many other AWS services for our infrastructure (code pipelines, security, search, and networking), our core application consists only of the services listed above, thus providing a consistent and simplistic architecture across many microservices.
Through tagging of resources, we also have much greater insight into the cost to run our application. As a business with profitability as a core value, we are constantly monitoring spikes in usage to identify how changes in our organization and application affect our costs. With the ability to easily monitor costs at the granularity of a single Lambda, profitability can be a business tenet that our entire product development team feels empowered to be a part of.
SaaS Factory: What technical challenges did you face building this SaaS solution on AWS?
Chris Petrunik: One of the biggest hurdles was handling function failures, timeouts, and retries. Functions-as-a-service (FaaS)-based development can magnify the impact of these issues, but we’ve quickly identified runtime issues using Amazon CloudWatch metrics. We’ve also built native core logging and monitoring capabilities into our stack. Best practices for serverless-based development continue to evolve, and we plan to share some of our insights with the community as we learn more.
SaaS Factory: What support did AWS SaaS Factory provide your team?
Chris Petrunik: We knew we’d need the support of AWS and the SaaS Factory to make our goals achievable. While serverless technologies have been around for a while, we haven’t found many examples of enterprise applications designed specifically around the FaaS architecture style.
AWS has supported us by conducting many technical deep dives with our team on a wide range of topics, including DynamoDB and disaster recovery. They also provided high-level AWS overviews for our non-technical employees so that everyone can be a part of the conversation.
The most valuable part of the partnership has been their willingness to push us to think about long-term decisions earlier in our evolution. They have seen the difficulty in implementing principles like multi-tenancy in established products, and have worked with us to incorporate these capabilities into our architecture from the start.
SaaS Factory: What would you tell others planning to build a SaaS solution on AWS?
Chris Petrunik: Ridgeline uses a single cloud provider—AWS. We believe the depth in offerings and a close partnership will give us everything we need to deliver best-in-class applications to our customers.
To others who are just starting out, we would advise them to choose the right tool for the job. The opportunity cost of investing in a vendor-agnostic design early on is an expensive one. At Ridgeline, we decided the value AWS could provide outweighed this option. We’d suggest you find providers you trust and who have had success supporting companies in a similar space.
AWS provides an immense number of services, many of which we hope to leverage at Ridgeline over time. Our advice to others would be to keep your AWS “footprint” as simple as possible for as long as possible. Availability is key in the SaaS world and for other companies like Ridgeline, who live by compliance standards like SOCII, it is a requirement.
As Ridgeline builds out our disaster recovery processes and structures, the simplicity of our architecture has made what is normally a multi-year effort achievable with a fraction of the resources.
Many of the serverless functions we use (Amazon DynamoDB, AWS Lambda, and Amazon API Gateway) make our disaster recovery model extremely simple, as they were built with a multi-region deployment in mind. Ridgeline will continue to add on more AWS products as our needs become more complex, but we are glad we started simple.
All of these services share common themes: they will scale on-demand, are fault-tolerant, and you pay only for what you use. In the enterprise world, we support large, complex customers who rely on our systems to run their core business functions. Scalability and availability are two attributes customers pay a premium for and traditionally enterprise SaaS vendors have achieved this through overprovisioning resources.
This is expensive and often results in the vendor incurring high costs for idle and unused resources. Serverless technologies provide rapid scalability and stability across AWS Availability Zones and regions but without provisioned resources. We believe serverless technologies will have a dramatic impact on our ability to efficiently serve our customers.
SaaS Factory: What are your future plans with AWS?
Chris Petrunik: This is a huge, multi-year project requiring significant investment in infrastructure and technology, which is one of the reasons it’s so exciting to have AWS as a partner. Ridgeline will continue to add on more AWS products as our needs become more complex, but we are glad we started simple.
Learn More About AWS SaaS Factory
AWS SaaS Factory provides a comprehensive set of business and technical best practices, and engagement opportunities that empower and support independent software vendors (ISVs) building SaaS offerings on AWS.
AWS Technology Partners are encouraged to reach out to their account representative to inquire about engagement models and to work with the AWS SaaS Factory team.
Additional technical and business best practices can be accessed via the AWS SaaS Factory website.