In order to achieve interconnection and intercommunication of various hospital systems' data, and break through the "information island" problem, various integration platforms are constantly emerging, and integration engines play a crucial role as core components. However, in the process of platform construction and operation and maintenance, hospitals face many problems and pain points. How can these pain points be solved?

Pain Point 1: After the hospital builds the integration platform, how can it truly take over the platform, reflect the application effects, turn the data into its own, and achieve autonomy and control?

Odin Solution Key Points:90% API openness; multi-functional engine; embedded domestic and foreign medical standardization components; easy to use, low learning cost, easy to operate and develop.

After the vendor completes the implementation of the integration platform, the hospital often cannot control the platform. Since improving data application efficiency requires follow-up operation and development, the integration platform often becomes a black box with opaque data. The hospital needs to spend a lot of time, effort, and money to seek help from the vendor for operation and maintenance issues, and cannot truly take over the platform as intended.

The poor usability of the integration engine, high threshold for secondary development, and lack of some common functions are the main reasons why hospitals cannot take over the platform.

To address these pain points, Odin has carried out a lot of localized secondary development on the engine, added many functions and usability that are needed in China but not available in foreign counterparts; Odin engine is fully in Chinese interface, with pure web operation interface and visualization tools, making operation and maintenance simple and clear, reducing the learning curve; the product embeds Groovy dynamic script and shared script functions, with 90% API openness to help hospital technical staff quickly get started and conduct secondary development independently, so that the hospital can truly take over the platform. This achieves the autonomy and control of middleware products by users, and integration vendors can also avoid excessive services, thus achieving a win-win situation.

During the COVID epidemic, Odin's user - Taizhou Hospital in Zhejiang Province needed to quickly respond to the promotion of the Chinese national "Health Code". Odin engine has a drag-and-drop, graphical operation interface, achieving “what you see is what you get”, greatly reducing the workload of operators; it also embedded national medical standardization components such as CDA and HL7, meeting the technical requirements of different platforms and scenarios; the convenient medical information integration and exchange mode solves the interoperability of medical information; the data service project also provides online synchronous debugging and hot deployment functions, with efficient interface iteration capability. With the assistance of Odin engine, a skilled technical staff from the hospital's information department completed all service deployment independently within one hour, which would normally take 2-3 days under traditional interface mode.

In addition, during the construction of the integration platform, the technical staff of the integration vendors often spend a long time learning to use the integration engine, and the secondary development of the project is cumbersome and complex, and the later maintenance is not easy. If there is a replacement or discontinuity of technical staff, it is difficult to provide timely and effective technical support to the hospital. The ease of use and functionality of Odin engine and other features have also solved the above problems for integration vendors, achieving a win-win situation for both hospitals and integration vendors.

Pain Point 2: How can hospitals reduce the risk of business interruption caused by integration platform downtime?

Odin Solution Key Points: Application-level software cluster; high availability of core business under high concurrency; integrated supervision of multiple production servers.

The hospital's integration platform must run continuously 7x24 hours, and platform downtime causing the interruption of hospital-related businesses is an unbearable "pain" for the hospital. Nowadays most hospital integration platforms adopt cold backup or dual hot backup solutions, which can meet the most basic disaster recovery requirements. However, when problems occur, the cold backup solution requires manual switching to the backup machine, resulting in a long business interruption time; the dual hot backup solution, although it can achieve automatic switching, may also encounter situations where both the main and backup machines fail. Many hospital operation and maintenance personnel have reported that they sometimes cannot sleep well because they are worried about platform downtime causing business interruption, and when they receive an emergency call, they need to go to the site to solve the problem immediately.

To address the above pain point, Odin has launched a cluster solution developed based on the business characteristics of hospital platforms. This solution separates functions such as management monitoring, production services, and data storage, and uses load balancing to build a cluster system, achieving the effect of one configuration running in multiple locations with consistent data.

Different from the general operating system-level cluster, which is difficult to balance high availability and high utilization of hardware resources, Odin's application-level software cluster can distribute services to multiple servers for simultaneous operation, and can also be dynamically expanded. Even if a server fails, other servers can seamlessly take over the business, combined with the cluster architecture to enhance the overall system running capability and the efficient and reasonable dynamic task distribution mechanism provided by load balancing. This ensures that the integration platform can fully utilize hardware resources, and achieves high availability in high concurrency and high data throughput environments, allowing hospital operation and maintenance personnel to rest assured and sleep soundly.

Pain point 3: How can a hospital integration platform meet the data exchange of tens of millions of level with millisecond-level delay in high-concurrency environment?

Odin solution key points: Over 80% of resource elasticity utilization rate; proactive architecture elastic scaling; integration engine, ESB (Enterprise Service Bus), and ETL (Extract, Transform, Load) "Three in One"

Nowadays, hospitals are connected to numerous systems and interfaces with various types and docking methods, which brings a heavy burden to the integration platform in high-concurrency environments with large traffic. When using traditional integration engines, performance bottlenecks often occur, resulting in slow response times for interface requests, low transmission efficiency, interface lag, and even crashes.

The Odin engine has a proactive architecture, and the data service functions of the integration engine, ESB, and ETL work together, exerting their respective advantages. With the characteristics and advantages of load balancing, distributed containers, elastic scaling, and hybrid deployment, it achieves ultra-high-concurrency performance, fine-grained resource management for projects, and can achieve over 80% of resource elasticity utilization rate, fully tapping into resource efficiency.

According to statistics in 2019, more than 300 data exchange projects have been built on the Odin engine at Taizhou Hospital in Zhejiang Province, which connects more than 200 self-service terminals and dozens of hospital internal and external systems with the platform and integrates 40% of the core business processes of the hospital group. Now Taizhou Hospital has integrated over 400 projects and thousands of interaction terminals, and the engine has the capability to integrate more projects and support larger transaction volumes. Taizhou Hospital keeps up with the times, continuously strengthens the breadth and depth of integration.

Pain point 4: How can a hospital integration platform adapt to changes in new technologies and new scenarios?

Odin solution key points: Upgrade from "Three in One" to multifunctional; support new technologies, including Kafka, Hadoop, and the Internet of Things.

With the rise and application of new technologies and new environments such as 5G, containerization, the Internet of Things, and hybrid clouds, the integration platform must keep up with the times. If the integration platform cannot keep up with the development speed of new technologies, it will become a pain point for hospitals in the future, and may need to be rebuilt.

As a technology innovation enterprise that has received national innovation and growth scientific research subsidies from New Zealand, Odin's proactive architecture is not only aimed at hospitals' current pain points but also future-oriented. On the basis of the original "three in one" functions of the integration engine, ESB, and ETL, it has added powerful support for new technologies, including Kafka distributed streaming platforms, Hadoop big data applications, and various new technologies such as MQTT and XMPP, which are used for IoT application layer protocols. Odin will continue to provide support for hospitals' constantly changing new technology and new application scenario requirements.

Pain point 5: How can super-large hospital groups, medical consortia, and regional projects achieve scaled integration construction?

Odin solution key points: Odin NeXT, the latest container orchestration technology, cloud-native distributed cluster architecture.

Although the multifunctional engine of Odin can solve most of the pain points in the construction of hospital integration platforms, for super-large-scale medical institutions such as regional medical consortia and group hospitals, the advent of 5G has accelerated the process of medical information interconnection and intercommunication. The access of more devices with higher speeds, lower delays, and more devices means a surge in data processing volume. The performance and architecture of traditional integration engines cannot support the future integration platform's scaled construction needs for such large medical groups.

In mid-2019, the Odin NeXT (New generation eXchange Technology) distributed cloud-native data service platform was launched. Odin NeXT is specifically designed for the needs of super-large-scale data centers and large-scale cloud applications. It is based on the latest container orchestration technology, Kubernetes (K8s), and a PaaS-level cloud-native distributed cluster architecture that uses containerization technology to provide a high-availability, high-concurrency, high-performance, low-latency cloud platform. Its "three highs and one low" characteristics fully support the large-scale data interconnection needs of the 5G environment. At the same time, Odin NeXT supports public and private hybrid clouds, is oriented towards microservices architecture, achieves business data decoupling, and meets the needs of future large-scale medical institutions' scale integration construction through its elastic and intelligent unlimited expansion feature.

Conclusion

In conclusion, autonomy and controllability, high availability, high performance, and support for new technologies are the pain points in the construction of hospital integration platforms. Odin adopts the latest technology, continuously innovates, and carries forward the spirit of craftsmanship. In terms of functionality, it has achieved a four-level leap from the integration engine to "three in one" to "multi-function" to "multi-function +", and in terms of architecture, it has completed the transformation from a single machine cold backup to high availability (HA) to cluster and cloud-native architecture. Odin has both the innovative Odin multi-function engine that solves current problems in hospital institutions, as well as the cloud-native distributed cluster version of Odin NeXT, which is based on the latest technology such as K8s and helps to build a new generation of hospital information systems for the future 20-30 years.