PatientPing is a nationwide platform that keeps medical professionals abreast of patients’ treatment across facilities. With VividCortex , PatientPing engineers take ownership of their code - e.g. finding underperforming queries and making changes so they run up to 10x faster.
PatientPing is a nationwide care coordination platform that is reducing healthcare costs — and achieving explosive growth — by keeping medical professionals fully abreast of patients’ treatment across facilities. Doctors, hospitals, emergency rooms, outpatient clinics, and other network providers get real-time admissions and discharge notifications, or Pings, whenever and wherever patients receive care. PatientPing also gives care providers immediate access to critical patient history. As a result, the company’s ability to quickly and reliably manage vast quantities of complex data is of crucial importance to thousands of patients and providers every day. Building out a patient-centric platform to serve multi-layered healthcare networks across the US is a huge job. PatientPing’s engineers know that the continued success of their ground-breaking application depends in large part on how effectively it interacts with their MySQL database.
Engineers Take Ownership of Their Code with Database Monitoring
“MySQL Workbench is okay but relying on it to monitor our database performance was like the old elephant joke: each blindfolded person who touches a different part thinks it’s a different animal,” said Heidi Schmidt, Data Engineer for PatientPing. “We needed a way to let our whole team see the complete picture at all times.”
To address this problem, VP of Engineering Ryan Scharer asked his team to find a tool that matched their values: "We empower developers to understand exactly how their code behaves in production, without being mediated through a special class of operators or DBAs, so they can take responsibility for the full lifecycle of their solution. We have such tools in place for monitoring application performance and hardware resources, but the database was essentially a black box. VividCortex solved this for us. Every team now has access to real-time data on how each of their queries is impacting the system, and they can tune their own code in response."
Preventative Care Pays Off: Key Results
When PatientPing implemented VividCortex on its core systems, data engineers immediately saw that several of their applications were dropping errors and that key warning signs were not being bubbled up within the ELK stack (Elasticsearch, Logstash, and Kibana). Bringing previously unknown errors into full view allowed the PatientPing team to detect anomalies and anticipate problems so they could take preventative action.
- Training Program enabled dozens of engineers, data scientists and others to use VividCortex to monitor database performance
- Uncovered individual queries that were not performing as intended and made improvements so those queries run up to 10x faster
- Found/fixed a problematic query pattern to achieve double-digit reductions in memory use and stabilize at a new sustainable level to reduce risk of outages
PatientPing, a proven leader in healthcare information technology, is now a leader in database performance monitoring.
"The bottom line here at PatientPing is supporting customers by optimizing
every interaction between our app and database ...We’ve been able to use VividCortex to... take proactive measures to keep our database healthy!”"
—Heidi Schmidt, Data Engineer