Know before the phone rings.
A tube that has been waiting too long, a disk filling up, an application slowing down: Jade Monitoring shows it on a screen, before a phlebotomist or a biologist calls you.
Where is this tube, and since when?
Jade knows when a tube was received, when the analyser returned its result and when the biologist validated it. Jade Monitoring gathers these moments and lays them on the plan of your premises: a tube becomes a route, as on the screen above, and a place becomes a waiting time.
Each application adds what it sees of the tube
- Jade
The reception of the tube, the results of its tests and the biological validation.
- Jade Online and LysPatient
They can tell when the results are released to the patient: the journey can then be followed to the end.
- The rest of the suite
The plan accepts other sources: an application of the suite that sees the tube go by can add its moments.
Live
What is happening now: the tubes in progress, those past their target, the places under pressure.
Replay and statistics
Yesterday, or the last seven days: replay how busy the laboratory was and compare each segment with its target.
From the plan of your premises to following a tube
- Draw
Import the plan of your premises as a background, then outline the places: reception, analysers, workstations, storage.
- Link
Attach each place to what Jade knows: an analyser, reception, validation.
- Set the targets
Give each place its target stay and its critical threshold, in minutes.
- Follow
Scan a barcode: the route of the tube appears, with the targets it exceeded.
- No patient data on the plan: only the tube number appears
- Searching for a tube is a separate permission, reserved for the roles you choose
- The plan reads Jade and never writes to it
- Median, P95 and overruns for each segment of the route
Are the servers holding up?
An agent on each Jade server reports processor, memory, disks and services. Written in Rust, it fits in a single file of about 5 MB, with no Java and nothing else to install.
A single file
A native executable of about 5 MB, installed as a Windows service. No Java, no runtime, nothing to maintain alongside it.
The agent makes the call
The agent listens on no port: it opens the connection to Jade Monitoring itself.
Isolated machines too
A server with no direct access goes through a neighbour's agent, which acts as a gateway and relays without reading anything.
Updated remotely
Every version is signed. The rollout starts with one pilot server, continues in waves, and rolls back on its own if an agent fails to check in again.
From Windows Server 2008 R2 onwards
The servers you cannot replace yet are monitored like the others.
It measures, it does not pry
Process names and usage figures. Never a command line, never the content of a file.
What the agent reports
- Processor and memory, per server and per service
- Every disk volume, with its fill trend
- The state of Windows services and the ports they listen on
- The processes using the most memory and processor
- Automatic restart of a stopped service: you are only notified if it fails
- Warning and critical thresholds, global or specific to a server
Are the applications doing well?
An application knows better than anyone how it is doing. It publishes its indicators through a health API, Jade Monitoring reads them every minute and turns them into a single state per application. Jade already speaks this language; any application that exposes the same API is monitored in the same way.
- The application publishes
A few indicators, each with its value and its state: running, pending tasks, age of the oldest one.
- Jade Monitoring reads
Every minute, on each server where the application runs.
- The state travels up
From the indicator to the instance, from the instance to the server, from the server to the application. The most serious state wins.
- You are notified
By e-mail, according to your rules: who, at what times, after what delay.
Your turn: click an indicator to change its state, and follow it up to the application.
- ServerSRV-JADE-APP01OKInstanceST-1:8080InstanceST-2:8081
- ServerSRV-JADE-APP02OKInstanceST-1:8080InstanceST-2:8081
- ServerSRV-JADE-APP03OKInstanceST-1:8080InstanceST-2:8081
The most serious state wins
One health API, the same for all
Jade exposes it today. Jade Online, LysPatient or one of your own applications: as soon as an application publishes its indicators in this format, it is monitored here, with the same aggregation and the same alerts.
Jade LISMonitored today
Jade OnlineAs soon as the API is exposed
LysPatientAs soon as the API is exposed- Your applicationAs soon as the API is exposed
First-hand indicators
They come from the application itself: its state, its pending tasks, the age of the oldest one.
One model per application
You describe once where to read its health. The model then applies to every server it runs on.
A view per application
One card per application: the servers it runs on and, when it is in trouble, the cause shown in plain words.
Alerts sent to the right people
E-mail rules per scope, with a time window and a delay before sending.
Underneath, a workflow engine
The core of Jade Monitoring is an engine that fetches a piece of data, transforms it, then displays it or raises an alert. It is what feeds the laboratory floor plan. Lysoft delivers it configured: you do not have to write a workflow to use it.
- Fetch
A query in Jade, an HTTP call, the latest reading from an agent, a certificate.
- Transform
Filter, join, compare with the previous value, aggregate over a time window.
- Display or alert
A dashboard tile, an alert, an e-mail, the restart of a service.
A sample workflow: the median TAT of the laboratory
And the dashboard it feeds
The outlined tile comes from the workflow above. The others are built the same way.
- Tiles laid on a grid: counters, curves, bars, lists
- A period and a site chosen at the top of the page, applied to the whole dashboard
- A mode for the laboratory wall screen, and a PDF export
- A dashboard you can embed in a page of your intranet, through a signed link
Advanced featureFor teams who want to build their own
The workflow editor is open to your IT team. Nothing requires you to open it: here is what they would find in it.
- Data
- Oracle and PostgreSQL SQL, HTTP, agent, TLS certificate, files and folders
- Transformation
- Mapper, join, condition, sliding aggregation, change detection
- Output
- Tile, alert, outgoing webhook, service restart
- Timing
- Scheduling, on-demand runs, reusable sub-workflows
Frequently asked questions
Does anything need to be installed on the Jade servers?
One agent per server: an executable of about 5 MB, installed as a Windows service. The health of an application can also be read directly by Jade Monitoring, without an agent.
Which systems are supported?
The agents run on Windows, from Windows Server 2008 R2 to current versions. Jade Monitoring itself is installed on Windows or Linux, in your infrastructure.
Does the plan show patient data?
No. The plan only knows the tube number, and searching for a tube is a separate permission that you grant to the roles of your choice.
Do we have to write workflows?
No. The plan, server monitoring and application health work without touching them. The editor is there for teams who want their own dashboards.
How are we notified?
By e-mail, according to your rules: which events, for which servers, at what times, after what delay. A service that restarts on its own wakes nobody up.
Which applications can be monitored?
Any application that exposes the health API. Jade does today; Jade Online or LysPatient are monitored in the same way as soon as they expose it.
See Jade Monitoring running
We show you the plan, the server list and application health on a demonstration environment, then we look together at what it takes to connect it at your site.
Jade Monitoring is for laboratories equipped with Jade.
The screens on this page are animated mock-ups, on sample data.
Request a Demo
Fill out the form below and our team will contact you within 24 hours to schedule your personalized demonstration.