Encontrar

Article
· Apr 7, 2023 2m read

Autoscaling IRIS Workloads. My adventure with IKO, HPA, and Traffic Cop

This week I was able to demo a proof of concept for our FMS interface on traffic cop architecture to my team. We are working on modernizing an Interoperability production running on mirrored Health Connect instances. We deploy IRIS workloads on Red Hat OpenShift Container Platform using InterSystems Kubernetes Operator (IKO). We can define any number of replicas for the compute stateful set where each compute pod runs our Interoperability production. We introduced Horizontal Pod Autoscaler (HPA) to scale up the number of compute pods based on memory or CPU utilization. But IKO scaled down because it wanted to keep the defined replicas. When compute pods receive shutdown signal while they are busy, messages in queues do not get processed right away. 
We are transitioning to "Traffic Cop" architecture to enable us to autoscale our workloads. Instead of deploying interoperability production on multiple compute pods, we deploy it on a mirrored data pod which functions as a traffic cop. We will create more REST interfaces where the message processing happens on stateless compute pods which can be deployed without IKO and no interoperability production will be on stateless computes. 
Compute and webgateway containers run as sidecar containers in one pod where webgateway receives requests to be processed in its paired compute. 
Along the way I have created a REST service running on our stateless compute pods which was started with a Swagger API specification. InterSystems IRIS API Management tools generated the code for the REST interface instead of manually coding it.

Discussion (0)1
Log in or sign up to continue
Question
· Apr 5, 2023

How to know which users did patient searches on clinical viewer

Hi community,

I would like to know how to know which users did patient searches on clinical viewer.

I tried checking the audit database but I cannot seem to know if it's logged there and under which event.

 

Please advise how to go about this.

 

Regards,

Ephraim Malane

2 Comments
Discussion (2)1
Log in or sign up to continue
Announcement
· Mar 24, 2023

[Video] Git Source Control for InterSystems IRIS Interoperability with Docker and VSCode

Hi Developers,

Often we create and edit InterSystems IRIS Interoperability solutions via a set of UI tools that is provided with IRIS. But it is sometimes difficult to setup the development environment to handle changes we make in the UI to source control.  

This video illustrates how git-source-control helps with source control Interoperability components while changing it in the UI.

⏯ Git Source Control for InterSystems IRIS Interoperability with Docker and VSCode

Add these two lines in your iris.script during docker build:

zpm "install git-source-control"
do ##class(%Studio.SourceControl.Interface).SourceControlClassSet("SourceControl.Git.Extension")

And Interoperability UI components will start working with git. 

Example application.

Discussion (0)2
Log in or sign up to continue
Please note that this post is obsolete.
Announcement
· Mar 20, 2023

Sneak Peek (at least for this week)

 Hi Community:

As you may know, I am the person who does the program and scheduling for our Global Summit sessions, so I can give you a sneak peek at our agenda before it is published.

I'm happy to report that we will have two special, four-hour workshops on Sunday, June 4, which is the day you arrive to register in Hollywood, Florida.

One is the return of the standing-room only 2022 workshop with our FHIR guru, Russ Leftwich - "Getting to Know FHIR: The Best Explanation of FHIR They've Ever Heard." Needless to say, we've reserved a bigger room.

Plus, we are planning eight related breakout sessions:

  1. "FHIR to IntegratedML: Can You Get There from Here?'
  2. "Building a FHIR Façade" 
  3. "How to Customize Your FHIR Server"
  4. "Clinical Alerts and Notifications on FHIR: Putting the Healthcare Action Engine into the Workflow"
  5. "Performing Advanced FHIR Validation"
  6. "Clinical Research Data Pipeline using FHIR and OMOP"
  7. "FHIR in the Cloud"
  8. "Introducing Bulk FHIR Export"

Our second workshop in an introduction to containers -- "Working with Containers and InterSystems Technology" - led by Derek Robinson from our online learning team. It will prepare you for these two related sessions

  1. "The Container Lifecycle: When is Your App Ready to Accept Work?"
  2. "Can you Autoscale This?  Lessons from the Field on Kubernetes"

Register TODAY for Global Summit 2023 to take advantage of Early Bird Prices. (You will be able to register for sessions, including the workshops, soon.)

Discussion (0)1
Log in or sign up to continue
Article
· Mar 16, 2023 6m read

Kinds of properties in IRIS

InterSystems IRIS has quite a few different kinds properties. Let’s put them in order so that they make better sense.

First of all, I would divide them into categories:

  • Atomic or simple properties (all those %String, %Integer, %Data and other system or user datatypes)
  • References to stored objects
  • Built-in objects
  • Streams (both binary and character)
  • Collections (which are divided into arrays and lists)
  • Relationships (one-many and parent-children)

Some of these kinds of properties are quite straightforward. For example, atomic properties:

Property Name As %Name;
Property DoB As %Date
Property Age As %Integer

They are also easily created using Studio Wizard:

The concepts of references to stored objects and built-in objects are also quite easy to grasp. Basically, if the class of an object you’re trying to use as a property extends %Persistent, then it’s a reference to a stored object. If the related class extends %SerialObject – then it’s a built-in object, because such objects can’t be stored themselves, only inside other objects. And in a class definition they look exactly the same:

Property Human as Sample.Human;

To create this kind of property, on the second step of the Wizard in Studio enter the name of the class:

Streams are also quite easy to explain. You have a big chunk of unstructured data, be it binary or character, and here is your stream. Can be anything – audio, video, document, text etc. Depending on what type of data you want to store, you may choose to use Binary or Character Stream. Another defining characteristic of a stream is the place where you want to store it – on disk (takes less space but accessible from your OS unless you set up access rules for the folder) or in the database (secure, no unauthorized access, takes more space). To work with streams, use classes from package %Stream. For example:

Property Log As %Stream.GlobalCharacter;
Property Instruction As %Stream.FileCharacter(LOCATION = "D:\Temp");
Property Doc As %Stream.FileBinary;
Property Audio As %Stream.GlobalBinary;

In this case input the classname of the stream you wish to use:

Then there are two types of collections:

  • Lists – a collection where each element is defined by its position in the collection
Property FavoriteColors as list of %String;
  • Arrays – key-value pairs, where value is defined by its key that is set by a user
Property Contacts as array of %String;

Both collections can be of simple types or of objects. And when we’re talking about a collection of objects, the objects inside collection aren’t “aware” that they are inside any collection. Meaning that it’s a one-way link.

When working with arrays, it’s necessary to remember that both key and value should be new and useful piece of info. There is no point of making an integer key that will imitate a position in a list. In the example above, key of the Contacts array is a name of the type of the contact, e.g. “Facebook”, “LinkedIn”, “Twitter” etc and the value is a link to a personal page.

Under the hood, when working with a list, you’re working with the class %Collection.ListOfDT for a list of simple types, %Collection.ListOfObj when working with a list of objects and %Collection.ListOfStream when dealing with a list of streams.

The same is true for arrays. %Collection.ArrayOfDT is used when working with an array of simple datatypes, %Collection.ArrayOfObj – when working with an array of objects and %Collection.ArrayOfStream – when working with an array of streams.

For the collection on the second step of the Wizard in Studio choose the second option “The collection of type” 

and then specify the type:

And probably the most challenging kind of the property for people who switch from relational databases – relationship. The relationship is a two-way one-to-many link between two stored objects. In relational DBMS people are taught that they need an additional table to store the foreign keys to two other tables to realize one-to-many link of independent entities. For example:

table Invoice – table Product – table Invoice_Product

There is no need for an additional table/class to do this in IRIS. If there’s no need to often query the information about all the invoices in which the exact Product is listed you can make products as an array or list in an invoice. In this case you will have to manually ascertain logical integrity so that your invoice doesn’t reference products that are no longer in a DB.

To automatically check for the logical integrity of data in this case you can use a relationship between these two classes: Invoice, Product. There are two types of relationships:

  • Parent-children – it’s a dependent link between objects of two different persistent classes. It means that the dependent object can’t exist without the main object. For example, let’s say that a chapter in a book can’t exist without a book. This will be an example of parent-children relationship, where a book is a main object and chapter is a dependent object and if we delete a book all the chapters from this book will be deleted as well.
  • One-many – it’s an independent link between objects of one or two persistent classes. It means that both objects can exist and make sense one without the other and if you try to delete the aggregating object, you’ll get an error saying that you first have to unlink all the linked objects. If you try to delete the linked objects, they will disappear from the aggregating object. Let’s take our invoice and products as an example. We have many products in an invoice. If we try to delete an invoice, we first need to delete all the products from our invoice. If we delete a product, it will disappear from the invoice.

Since it’s a two-way link you need to define the properties in both classes. For example in a class Sample.Invoice you will have the definition:

Relationship Products As Sample.Product [ Cardinality = many, Inverse = Invoice ];


Note that the property is called Relationship and that it has two characteristics:

Cardinality = manymeaning that there are links to many objects inside this property

Inverse = Invoicethis is the name of the property on the other side of the relationship

At the same time in the other class (or it can be in the same class for one-many relationship) there should be the mirror property:

Relationship Invoice As Sample.Invoice [ Cardinality = one, Inverse = Products ];

Here cardinality “one” means that in this property there is a link only to one object.

To create a relationship using Wizard in Studio just choose on the second step Relationship:

Then choose the correct cardinality for the property in the class and fill in the name of the related class and property in it:

For the example with the book with chapters the properties would look as follows.

In a class Sample.Book:

Relationship Chapters As Sample.Chapter [ Cardinality = children, Inverse = Book ];

In a class Sample.Chapter:

Relationship Book As Sample.Book [ Cardinality = parent, Inverse = Chapters ];

This is the end of a brief overview of different kinds of properties present in IRIS. Hope it makes it clearer.

4 Comments
Discussion (4)2
Log in or sign up to continue