Eviden is a next-generation technology leader in data-driven, trusted and sustainable digital transformation. Backed by a strong portfolio of patented technologies and global leadership positions in advanced computing, security, AI, cloud and digital platforms, Eviden offers deep expertise across industries in over 47 countries. Eviden is an AWS Premier Partner and brings together 47,000 world-class talent to extend the possibilities of data and technology across the digital continuum, for current and future generations. Eviden is part of the Atos Group, which has annual revenues of over €5 billion.
We are committed to upskilling our employees and helping develop the next generation of cloud-centric talent. While the foundational knowledge you gain through training and certification is important, nothing beats real-world experience. We supplement in-person learning with hands-on opportunities, including Immersion Days, Gamedays, and working with AWS DeepRacer.
AWS DeepRacer allows users to train reinforcement learning models on the AWS Cloud and race them on a virtual track. Unlike traditional programming, where you define a desired output, AWS DeepRacer lets you define rewards for specific behaviors, such as going faster or staying in the center of the track. This hands-on experience gives learners a great opportunity to interact with the AWS Management Console and develop Python-based reward functions, building valuable skills around cloud-centric technologies and machine learning (ML).
To improve the event experience and streamline the management of the global AWS DeepRacer Series, Eviden adopted the open source AWS DeepRacer Event Manager (DREM) solution. In this post, we discuss the benefits of DREM and the experience for racers, event staff, and spectators.
Introducing AWS DeepRacer Event Manager
AWS DeepRacer Event Manager is an innovative application that Eviden deployed within its AWS environment. Comprised of AWS Cloud-centric services, DREM is designed to simplify the process of hosting in-person AWS DeepRacer events while providing a more engaging and immersive experience for both participants and spectators.
Building on their experience hosting AWS DeepRacer events in the UK, Eviden wanted to expand this exciting initiative globally. By adopting the DREM solution, Eviden's experienced event staff in the UK were able to seamlessly support their first-time AWS DeepRacer event hosts in locations such as Bydgoszcz, Paris, and Pune. The DREM solution allows Eviden to seamlessly configure and manage global AWS DeepRacer events. Within the platform, AWS DeepRacer cars at each event venue are registered, allowing remote configuration and model uploading through AWS Systems Manager. Additionally, a Raspberry Pi device is registered at each venue to act as an integrated timing solution. Using DREM's data-driven racing capabilities, this device captures and reports key performance metrics for each racer, such as best lap time, average lap time, and total laps completed.
Additionally, the tight integration of DREM’s timing solution and the event’s streaming overlay enabled Eviden to deliver a much more engaging experience to both in-person attendees and remote viewers, keeping everyone well-informed and immersed in the action throughout the entire event.
The following diagram shows the DREM architecture and components:
Racer Experience
For racers, the DREM experience begins with registration, where they are encouraged to upload their AWS DeepRacer model prior to the event. Authentication is handled seamlessly by Amazon Cognito, with a local Amazon Cognito identity store out of the box, as well as the flexibility to integrate with corporate identity provider solutions if desired. Once registered racers upload their model, DREM automatically scans for suspicious content and quarantines it if necessary before making the validated model available for competitive racing.
Event staff experience
For event staff, the DREM solution greatly simplifies the process of running an AWS DeepRacer competition. User management, model uploading, and naming conventions are all handled seamlessly within the platform, eliminating potential confusion regarding model ownership. To further enforce the integrity of the competition, DREM applies an MD5 hash to uploaded models, preventing unauthorized sharing of models between racers. Additionally, the DREM interface makes it extremely easy and efficient to upload multiple models to an AWS DeepRacer car, providing a much better experience compared to the car's native graphical UI. DREM also simplifies the management of the AWS DeepRacer car fleet and Raspberry Pi timing devices, allowing event staff to remotely remove models, restart AWS DeepRacer services, and even print user-friendly labels for the cars, allowing racers to easily connect using the provided tablets.
To further streamline the event management process, DREM offers pre-built scripts that allow devices to be seamlessly registered. These devices can then be fully managed remotely. Timekeeping functions are handled automatically within DREM using a Raspberry Pi device, pressure sensor and pressure sensor trimmings, either through custom DIY modifications or using the excellent Digital Racing Kings board. The DREM timekeeping system is a significant improvement over the alternative solution Eviden has used in the past. It captures vital race metrics such as time remaining, all lap times and fastest lap. Additionally, the system also provides the option to invalidate a lap if, for example, the car goes off course. Once a racer has completed their run, the data is securely stored in DREM and the leaderboard is automatically updated to reflect the latest results.
Audience Experience
As a global systems integrator, Eviden was determined to deliver a truly amazing experience for not only on-site attendees of the AWS DeepRacer Finals, but also for those watching the event on-site, remote viewers watching the race using a proxy, or those simply interested in watching the competition unfold. To achieve this, Eviden leveraged the seamlessly integrated streaming overlay and leaderboard capabilities of the DREM solution, which ensured that all attendees, both in-person and online, were fully engaged and aware of their current standings throughout the entire event.
In previous AWS DeepRacer events, participants had to rely on someone on-site to verbally communicate lap times and time remaining in the race. But with DREM, both racers and spectators have instant access to all the important timing information, keeping everyone up to date. This is especially beneficial for remote participants, who can now clearly see which cars are on the track and follow the progress of other racers. A streaming overlay dynamically updates to show top leaderboard positions.
In addition, DREM will provide a dedicated web page displaying the full leaderboard, which can be conveniently displayed on a screen in the event space or allowing remote participants to follow the competition's progress from other locations throughout the day.
The significant improvements to the event experience were clearly reflected in the feedback received from this year's attendees.
- “It was great to be able to participate remotely.”
- “This year has seen a huge improvement in remote interaction between teams.”
- “The physical event was fantastic, well attended and very smoothly run.”
- “Even though we were participating remotely, it was still a lot of fun.”
- “The event itself was fantastic, each stage was well planned and organised. The atmosphere in the venue exceeded my expectations and it was a really great event to be a part of.”
Well-designed
The DREM solution is meticulously designed with a well-architected approach. From an event organizer's perspective, peace of mind is crucial knowing that DREM is protected using AWS WAF, Amazon CloudFront, and AWS Shield Standard, and user management is seamlessly handled by Amazon Cognito. Additionally, the platform's role-based access control (RBAC) is managed through AWS Identity and Access Management (IAM), enforcing least privilege policies for added security.
The DREM solution is built entirely using AWS Cloud-centric technologies, delivering inherent performance efficiency and reliability. When Eviden is not actively hosting an event, the DREM environment has minimal ongoing activity, consisting primarily of standing items such as AWS WAF rules, CloudFront distributions, Amazon Simple Storage Service (Amazon S3) buckets, Amazon DynamoDB tables, and Systems Manager fleet configurations. However, during an event, DREM seamlessly scales to dynamically meet the demands of your event hosting requirements using AWS Lambda, Amazon EventBridge, AWS Step Functions, and other serverless services.
Because the DREM solution is well-designed, the platform is incredibly cost-effective. If you're not actively hosting events, your DREM environment will incur minimal costs of around $6-8 per month, most of which comes from AWS WAF protection. During events, costs will increase depending on the number of users and models uploaded, but typically only rise to around $15 per month. To further optimize ongoing costs, DREM incorporates measures such as Amazon S3 lifecycle policies that automatically delete uploaded models after two weeks.
Conclusion
Are you interested in taking your own AWS DeepRacer event to the next level and provide a more engaging experience for your attendees? We invite you to explore the AWS DeepRacer Event Manager solution and see how it can transform your event management process.
To get started, visit the GitHub repository to learn more about the solution's features and architecture, and contact the Eviden team or your local AWS Solutions Architect to discuss how DREM can be customized to fit your specific event requirements.
Don't miss the opportunity to take your AWS DeepRacer journey to the next level: explore DREM and join us at an upcoming AWS DeepRacer event today.
About the Author
Satya Paduchuri Sathya is a Senior Partner Solutions Architect (PSA) at Amazon Web Services. Sathya helps partners run optimized workloads on AWS, build and develop their cloud practices, and develop new services.
Mark Ross Mark is the Chief Architect at Eviden and has specialized in AWS for the past 8 years, earning and maintaining all AWS certifications since 2021. Mark is passionate about helping customers build, migrate, and leverage AWS. Mark has built and grown a large AWS community within Eviden.
