- Industry
- Health and life sciences
- Organization
- Mentech
- Reon’s role
- Co-founder and CTO
The business need
Mentech's HUME platform processes sensor data from wearables in real time. As the number of users grew, the small IT team faced stability problems: an expensive external SQL database filled up quickly, and backpressure caused system errors.
What Reon did
- Reon worked with AWS partner DoiT to optimize Mentech's cloud platform.
- The platform moved from a single monolithic database to horizontally scaling AWS services, each matched to a type of data.
- HTTP/1.1 was replaced with HTTP/2 to deal with backpressure.
- An AWS Well-Architected Framework Review followed, and Mentech added database backups and multi-factor authentication.
Technology and approach
- Amazon Timestream for real-time data
- Amazon S3 for long-term storage
- Amazon Aurora for short-lived subsets of data
- Amazon RDS for metadata
- HTTP/2
Outcome
- Database costs fell by 60% while the user base grew fivefold.
- System reliability improved by 30%, with fewer tickets and errors.
- Overall AWS hosting spend fell by 10%.
- With DoiT's support, issues are fixed twice as fast.
Decisions this experience speaks to
- Our cloud costs are growing faster than our users. How do we optimise the platform?
- How do we combine wearables, sensor data and machine learning into a reliable clinical product?
Related services and use cases
Sources
This case describes Reon Smits’s own work before Grand Rounds, as Reon’s profile and the public sources above describe it. It is not a Grand Rounds engagement and not a promise of the same result for another company.