Welcome to the second entry in our blog series, “Enterprise Integration: Considerations for Healthcare Organizations”. In our first post, we identified a list of top considerations for organizations with large-scale, high volume deployments based on feedback from our customers. For this week’s installment, we will take a closer look at one of those key considerations: High Performance.
Healthcare organizations today are faced with integrating an increasing number of systems, applications, and devices, both internally and externally. As a result, these organizations must be able to process much larger volumes of data and do so in a reliable and predictable manner. Not to mention in a cost-effective fashion as well.
The need for a high performance integration engine may seem obvious when it comes to enterprise integration considerations, but it is surprisingly often overlooked. At a minimum, organizations who are evaluating the performance of an integration engine, should consider the following:
- Message Throughput – Can the integration engine handle your organization’s current message volume? An enterprise integration solution must be capable of consistently handling your organization’s peak message volume without any performance degradation.
- Scalability – Is the integration engine capable of scaling with your organization? As your organization grows, the messaging traffic (both in size and volume) your integration engine must handle will significantly increase.
- Hardware Requirements – What are the hardware costs required to meet your organization’s performance needs? Many integration engines can process high volumes of messages given powerful enough hardware. It is important to take the hardware costs into account, particular as your organization grows, as they vary greatly between integration engines.
Iguana 6 Performance Benchmarks
We recently (November 2017) conducted a series of tests to measure the Iguana’s performance for a variety of interface types, across three different server configurations.
In each test, Iguana ran multiple channels, where each channel was sent random traffic. The volume of traffic sent to each channel was randomized but capped at the numbers marked IMPS (input messages per second). In these tests, the number of channels and IMPS were adjusted until the highest sustainable performance was reached. In other words, the instance could handle traffic at these levels indefinitely without a message backlog.
Performance numbers listed in messages per second (MPS) represent outbound traffic that Iguana sent in response to the input traffic.
The tables below demonstrate Iguana’s performance results from three common integration scenarios. To view the rest of the results from our benchmark tests, please download the full PDF from the link below.
Download: Iguana 6 Performance Benchmarks (PDF)
HL7 Scrubber
HL7 v2 messages arrive over LLP, undergo simple text manipulation, and depart over LLP.
HL7 Filter
HL7 v2 messages arrive over LLP and are parsed into data trees. Fields are remapped, trees are serialized into HL7 v2 and depart over LLP.
Web Service
Requests arrive as JSON over HTTP. Each is parsed into a node tree, serialized as HL7 v2, and departs over LLP.
The results from our performance benchmarks clearly demonstrate that:
- Iguana can easily handle the high message volume that even the largest healthcare organizations face without generating a backlog of messages.
- Iguana operates at a high level on moderate hardware providing organizations with cost-effective scalability options.
Iguana In The Real World
Of course, performance must be proven outside of a testing environment. Iguana is used by healthcare organizations of all sizes and specialties. From some of the largest laboratories in the U.S. who process tens of millions of lab tests every year to software vendors handling millions of messages per day. For example, a large healthcare SaaS provider in the U.S. uses Iguana to process over 7 million HL7 messages from thousands of clients sites every day.
What do you think of the results from our performance benchmarks? Feel free to share your thoughts in the comment section.
As usual, please get in touch with us if you’re interested in learning more about Iguana’s capabilities or how healthcare organizations are using Iguana.
Next up in our “Enterprise Integration Considerations” blog series: High Availability.
Enterprise Integration Series